Authors: Luisa Di Paola, Micol De Ruvo, Paola Paci, Daniele Santoni, and Alessandro Giuliani.
Further information:
http://www.lswn.it/en/chemistry/news/protein_contact_networks_emerging_paradigm_chemistry
nternational Journal of Engineering Research and Development is an international premier peer reviewed open access engineering and technology journal promoting the discovery, innovation, advancement and dissemination of basic and transitional knowledge in engineering, technology and related disciplines.
Determining the Efficient Subband Coefficients of Biorthogonal Wavelet for Gr...CSCJournals
In this paper, we propose an invisible blind watermarking scheme for the gray-level images. The cover image is decomposed using the Discrete Wavelet Transform with Biorthogonal wavelet filters and the watermark is embedded into significant coefficients of the transformation. The Biorthogonal wavelet is used because it has the property of perfect reconstruction and smoothness. The proposed scheme embeds a monochrome watermark into a gray-level image. In the embedding process, we use a localized decomposition, means that the second level decomposition is performed on the detail sub-band resulting from the first level decomposition. The image is decomposed into first level and for second level decomposition we consider Horizontal, vertical and diagonal subband separately. From this second level decomposition we take the respective Horizontal, vertical and diagonal coefficients for embedding the watermark. The robustness of the scheme is tested by considering the different types of image processing attacks like blurring, cropping, sharpening, Gaussian filtering and salt and pepper noise effect. The experimental result shows that the embedding watermark into diagonal subband coefficients is robust against different types of attacks.
High Capacity Robust Medical Image Data Hiding using CDCS with Integrity Chec...IDES Editor
While transferring electronic patient report (EPR)
data along with corresponding medical images over network,
confidentiality must be assured. This can be achieved by
embedding EPR data in corresponding medical image itself.
However, as the size of EPR increases, security and
robustness of the embedded information becomes major issue
to monitor. Also checking the integrity of this embedded data
must be needed in order to assure that retrieved EPR data is
original and not manipulated by different types of attacks.
This paper proposes high capacity, robust secured blind data
hiding technique in Discrete Cosine Transform (DCT) domain
along with integrity checking. A new coding technique called
Class Dependent Coding Scheme (CDCS) is used to increase
the embedding capacity. High imperceptibility is achieved by
adaptively selecting the efficient DCT blocks. Even a slight
modification of stego image in embedded region as well as in
ROI (Region of Interest) can be detected at receiver so to
confirm that attack has been done. The embedding scheme
also takes care of ROI which is diagnostically important part
of the medical images and generates security key
automatically. Experimental results show that the proposed
scheme exhibits high imperceptibility as well as low
perceptual variations in Stego-images. Security and
robustness have been tested against various image
manipulation attacks.
The Effectiveness of PIES Compared to BEM in the Modelling of 3D Polygonal Pr...IDES Editor
The paper presents the effectiveness of the
parametric integral equation system (PIES) in solving 3D
boundary problems defined by Navier-Lame equations on the
example of the polygonal boundary and in comparison with
the boundary element method (BEM). Analysis was performed
using the commercial software “BEASY” which bases on the
BEM method, whilst in the case of PIES using an authors
software. On the basis of two examples the number of data
necessary to modeling of the boundary geometry, the number
of solved algebraic equations and the stability and accuracy
of the obtained numerical results were compared.
nternational Journal of Engineering Research and Development is an international premier peer reviewed open access engineering and technology journal promoting the discovery, innovation, advancement and dissemination of basic and transitional knowledge in engineering, technology and related disciplines.
Determining the Efficient Subband Coefficients of Biorthogonal Wavelet for Gr...CSCJournals
In this paper, we propose an invisible blind watermarking scheme for the gray-level images. The cover image is decomposed using the Discrete Wavelet Transform with Biorthogonal wavelet filters and the watermark is embedded into significant coefficients of the transformation. The Biorthogonal wavelet is used because it has the property of perfect reconstruction and smoothness. The proposed scheme embeds a monochrome watermark into a gray-level image. In the embedding process, we use a localized decomposition, means that the second level decomposition is performed on the detail sub-band resulting from the first level decomposition. The image is decomposed into first level and for second level decomposition we consider Horizontal, vertical and diagonal subband separately. From this second level decomposition we take the respective Horizontal, vertical and diagonal coefficients for embedding the watermark. The robustness of the scheme is tested by considering the different types of image processing attacks like blurring, cropping, sharpening, Gaussian filtering and salt and pepper noise effect. The experimental result shows that the embedding watermark into diagonal subband coefficients is robust against different types of attacks.
High Capacity Robust Medical Image Data Hiding using CDCS with Integrity Chec...IDES Editor
While transferring electronic patient report (EPR)
data along with corresponding medical images over network,
confidentiality must be assured. This can be achieved by
embedding EPR data in corresponding medical image itself.
However, as the size of EPR increases, security and
robustness of the embedded information becomes major issue
to monitor. Also checking the integrity of this embedded data
must be needed in order to assure that retrieved EPR data is
original and not manipulated by different types of attacks.
This paper proposes high capacity, robust secured blind data
hiding technique in Discrete Cosine Transform (DCT) domain
along with integrity checking. A new coding technique called
Class Dependent Coding Scheme (CDCS) is used to increase
the embedding capacity. High imperceptibility is achieved by
adaptively selecting the efficient DCT blocks. Even a slight
modification of stego image in embedded region as well as in
ROI (Region of Interest) can be detected at receiver so to
confirm that attack has been done. The embedding scheme
also takes care of ROI which is diagnostically important part
of the medical images and generates security key
automatically. Experimental results show that the proposed
scheme exhibits high imperceptibility as well as low
perceptual variations in Stego-images. Security and
robustness have been tested against various image
manipulation attacks.
The Effectiveness of PIES Compared to BEM in the Modelling of 3D Polygonal Pr...IDES Editor
The paper presents the effectiveness of the
parametric integral equation system (PIES) in solving 3D
boundary problems defined by Navier-Lame equations on the
example of the polygonal boundary and in comparison with
the boundary element method (BEM). Analysis was performed
using the commercial software “BEASY” which bases on the
BEM method, whilst in the case of PIES using an authors
software. On the basis of two examples the number of data
necessary to modeling of the boundary geometry, the number
of solved algebraic equations and the stability and accuracy
of the obtained numerical results were compared.
Genetic Programming based Image SegmentationTarundeep Dhot
Genetic Programming based Image Segmentation with Applications to Biomedical Object Detection. Published paper of our research work. Published at Genetic and Evolutionary Computation Conference (GECCO) 2009.
Transfer learning with multiple pre-trained network for fundus classificationTELKOMNIKA JOURNAL
Transfer learning (TL) is a technique of reuse and modify a pre-trained network.
It reuses feature extraction layer at a pre-trained network. A target domain in TL
obtains the features knowledge from the source domain. TL modified
classification layer at a pre-trained network. The target domain can do new tasks
according to a purpose. In this article, the target domain is fundus image
classification includes normal and neovascularization. Data consist of
100 patches. The comparison of training and validation data was 70:30.
The selection of training and validation data is done randomly. Steps of TL i.e
load pre-trained networks, replace final layers, train the network, and assess
network accuracy. First, the pre-trained network is a layer configuration of
the convolutional neural network architecture. Pre-trained network used are
AlexNet, VGG16, VGG19, ResNet50, ResNet101, GoogLeNet, Inception-V3,
InceptionResNetV2, and squeezenet. Second, replace the final layer is to replace
the last three layers. They are fully connected layer, softmax, and output layer.
The layer is replaced with a fully connected layer that classifies according to
number of classes. Furthermore, it's followed by a softmax and output layer that
matches with the target domain. Third, we trained the network. Networks were
trained to produce optimal accuracy. In this section, we use gradient descent
algorithm optimization. Fourth, assess network accuracy. The experiment results
show a testing accuracy between 80% and 100%.
Presented at Agile in Business conference at Bangalore. This presentation focuses on code metrics that can be used as lead indicators to effectively control and predict the application quality
ROBUST COLOUR IMAGE WATERMARKING SCHEME BASED ON FEATURE POINTS AND IMAGE NOR...csandit
Geometric attacks can desynchronize the location of the watermark and hence cause incorrect
watermark detection. This paper presents a robust colour image watermarking scheme based on
visually significant feature points and image normalization technique. The feature points are
used as synchronization marks between watermark embedding and detection. The watermark is
embedded into the non overlapped normalized circular regions in the luminance component or
the blue component of a color image. The embedding of the watermark is carried out by
modifying the DCT coefficients values in selected blocks. The original unmarked image is not
required for watermark extraction Experimental results show that the proposed scheme
successfully makes the watermark perceptually invisible as well as robust to common signal
processing and geometric attacks.
LaSiO 3 Cl:Ce 3+ ,Tb 3+ and Mg 2 TiO 4 :Mn 4+ : quantum dot phosphors for im...IJECEIAES
In this research, we focus on the solutions to enhance the lighting properties as well as the heat regulation of the white light-emitting diodes (WLEDs) with conventional phosphor and quantum dots (QDs). Although receiving lots of attention for being an innovative lighting solution with good color rendering index, the potentials of WLEDs conjugated with quantum dots (QDS), especially the QDs-phosphor mixed nanocomposites ones, are restrained due to the lacking performance in the aspects mentioned above. The crucial requirement to produce better WLEDs is finding solutions that improve the lacking aspects, therefore, through observing previous studies and applying advanced technique, this research suggest an effective and unique packaging configuration, in which the nanocomposites QDs-phosphor layer is set horizontally to the WLED. This novel packaging configuration allow WLED performance in terms of lighting and heating to reach it peaks. This is the first time four different types of WLEDs, singlelayer phosphor, dual-layer remote phosphor with yellow-red and yellow-green, and triple-layer phosphor, were simulated, utilized and compared in one study to decide the best WLED configuration. The results show that the triple-layer phosphor configurations improve the color rendering ability and lumen output better than the other configurations.
http://www.lswn.it/eventi/congressi/2012/tumori_neuroendocrini_gastropancreatici
Tumori neuroendocrini (Net) - Gastropancreatici (Gep) - La neoplasia che colpì Steve Jobs: rara e difficilissima da curare.
Progetto Minni - Sistema modellistico per le politiche di qualità dell'aria a...Le Scienze Web News
Il volume riassume le principali attività della Convenzione pluriennale 2008-2012 fra Ministero dell’Ambiente e della Tutela del Territorio e del Mare ed ENEA su “Sviluppo, verifica e nuove applicazioni del sistema modellistico MINNI a supporto delle politiche di qualità dell’aria nazionali e dei piani e programmi di risanamento della qualità dell’aria regionali”.
Genetic Programming based Image SegmentationTarundeep Dhot
Genetic Programming based Image Segmentation with Applications to Biomedical Object Detection. Published paper of our research work. Published at Genetic and Evolutionary Computation Conference (GECCO) 2009.
Transfer learning with multiple pre-trained network for fundus classificationTELKOMNIKA JOURNAL
Transfer learning (TL) is a technique of reuse and modify a pre-trained network.
It reuses feature extraction layer at a pre-trained network. A target domain in TL
obtains the features knowledge from the source domain. TL modified
classification layer at a pre-trained network. The target domain can do new tasks
according to a purpose. In this article, the target domain is fundus image
classification includes normal and neovascularization. Data consist of
100 patches. The comparison of training and validation data was 70:30.
The selection of training and validation data is done randomly. Steps of TL i.e
load pre-trained networks, replace final layers, train the network, and assess
network accuracy. First, the pre-trained network is a layer configuration of
the convolutional neural network architecture. Pre-trained network used are
AlexNet, VGG16, VGG19, ResNet50, ResNet101, GoogLeNet, Inception-V3,
InceptionResNetV2, and squeezenet. Second, replace the final layer is to replace
the last three layers. They are fully connected layer, softmax, and output layer.
The layer is replaced with a fully connected layer that classifies according to
number of classes. Furthermore, it's followed by a softmax and output layer that
matches with the target domain. Third, we trained the network. Networks were
trained to produce optimal accuracy. In this section, we use gradient descent
algorithm optimization. Fourth, assess network accuracy. The experiment results
show a testing accuracy between 80% and 100%.
Presented at Agile in Business conference at Bangalore. This presentation focuses on code metrics that can be used as lead indicators to effectively control and predict the application quality
ROBUST COLOUR IMAGE WATERMARKING SCHEME BASED ON FEATURE POINTS AND IMAGE NOR...csandit
Geometric attacks can desynchronize the location of the watermark and hence cause incorrect
watermark detection. This paper presents a robust colour image watermarking scheme based on
visually significant feature points and image normalization technique. The feature points are
used as synchronization marks between watermark embedding and detection. The watermark is
embedded into the non overlapped normalized circular regions in the luminance component or
the blue component of a color image. The embedding of the watermark is carried out by
modifying the DCT coefficients values in selected blocks. The original unmarked image is not
required for watermark extraction Experimental results show that the proposed scheme
successfully makes the watermark perceptually invisible as well as robust to common signal
processing and geometric attacks.
LaSiO 3 Cl:Ce 3+ ,Tb 3+ and Mg 2 TiO 4 :Mn 4+ : quantum dot phosphors for im...IJECEIAES
In this research, we focus on the solutions to enhance the lighting properties as well as the heat regulation of the white light-emitting diodes (WLEDs) with conventional phosphor and quantum dots (QDs). Although receiving lots of attention for being an innovative lighting solution with good color rendering index, the potentials of WLEDs conjugated with quantum dots (QDS), especially the QDs-phosphor mixed nanocomposites ones, are restrained due to the lacking performance in the aspects mentioned above. The crucial requirement to produce better WLEDs is finding solutions that improve the lacking aspects, therefore, through observing previous studies and applying advanced technique, this research suggest an effective and unique packaging configuration, in which the nanocomposites QDs-phosphor layer is set horizontally to the WLED. This novel packaging configuration allow WLED performance in terms of lighting and heating to reach it peaks. This is the first time four different types of WLEDs, singlelayer phosphor, dual-layer remote phosphor with yellow-red and yellow-green, and triple-layer phosphor, were simulated, utilized and compared in one study to decide the best WLED configuration. The results show that the triple-layer phosphor configurations improve the color rendering ability and lumen output better than the other configurations.
http://www.lswn.it/eventi/congressi/2012/tumori_neuroendocrini_gastropancreatici
Tumori neuroendocrini (Net) - Gastropancreatici (Gep) - La neoplasia che colpì Steve Jobs: rara e difficilissima da curare.
Progetto Minni - Sistema modellistico per le politiche di qualità dell'aria a...Le Scienze Web News
Il volume riassume le principali attività della Convenzione pluriennale 2008-2012 fra Ministero dell’Ambiente e della Tutela del Territorio e del Mare ed ENEA su “Sviluppo, verifica e nuove applicazioni del sistema modellistico MINNI a supporto delle politiche di qualità dell’aria nazionali e dei piani e programmi di risanamento della qualità dell’aria regionali”.
An Information Maximization approach of ICA for Gender ClassificationIDES Editor
In this paper, a novel and successful method for
gender classification from human faces using dimensionality
reduction technique is proposed. Independent Component
Analysis (ICA) is one of such techniques. In the current
scheme, a thrust is given on the different algorithms and
architectures of ICA. An information maximization ICA is
discussed with its two architecture and compared with the two
architectures of fast ICA. Support Vector Machine (SVM) is
used as a classifier for the separation of male and female
classes. All experiments are done on FERET database. Results
are obtained for the different combinations of train and test
database sizes. For larger
training set SVM is performing with an accuracy of 98%. The
accuracy values are varied for change in size of testing set and
the proposed system performs with an average accuracy of
96%. An improvement in performance is achieved using class
discriminability which performs with 100% accuracy.
Following the Evolution of New Protein Folds via Protodomains [Report]Spencer Bliven
Protein evolution proceeds through genetic mechanisms, but selection acts on biological assemblies. I define a protodomain as a minimal independently evolving unit with conserved structure. Protodomain rearrangements have minimal impact on biological assemblies, so they represent a valid evolutionary path through fold space.
This report is the written portion of my Candidacy Exam at University of California, San Diego. It discusses my current research in Philip Bourne's lab, as well as proposes research for my thesis over the next two years. Slides for the oral presentation are available at http://www.slideshare.net/sbliven/following-the-evolution-of-new-protein-folds-via-protodomains
Similar to Protein Contact Networks: An Emerging Paradigm in Chemistry (20)
XXXI convegno del Gruppo Nazionale di Geofisica della Terra Solida ProgrammaLe Scienze Web News
Dal 20 al 22 Novembre 2012 si svolgerà a Potenza il XXXI convegno del Gruppo Nazionale di Geofisica della Terra Solida.
Per informazioni: http://www.lswn.it/comunicati/stampa/2012/studiare_i_terremoti_come_il_meteo
La Settimana di Educazione allo Sviluppo Sostenibile vedrà quest’anno la sua settima edizione, con la partecipazione di centinaia di aderenti da ogni regione d’Italia che, sotto l’egida di questa Commissione, dedicheranno le loro iniziative al tema “Madre Terra: Alimentazione, Agricoltura ed Ecosistema”.
Incontri educativo-informativi sul tema della fertilità alla SapienzaLe Scienze Web News
Presentazione: Nel corso degli ultimi anni è stato registrato un preoccupante incremento delle affezioni acute e croniche della sfera riproduttiva. L'infertilità affligge ormai nel nostro paese oltre il 15% delle coppie che cercano di avere un figlio. Tutto ciò è correlato a comportamenti scorretti o dannosi acquisiti in età giovanile, dovuti ad una scarsa informazione e alla mancanza di prevenzione. La salute riproduttiva viene spesso trascurata e presa in considerazione solo quando le problematiche diventano eclatanti, con un ritardo che vanifica l’azione medica e si accompagna ad una crescita dei costi sanitari.
Numerose sono le cause e i fattori di rischio sui quali si può intervenire, così come le nuove terapie che consentono il raggiungimento di risultati precedentemente impensabili. Ciò ha reso necessaria la definizione di un programma di prevenzione finalizzato all’educazione alla salute riproduttiva come componente fondamentale della salute dell’individuo e della coppia e ad organizzare questi incontri informativi e di aggiornamento sulla diagnosi e cura dell’infertilità.
DIRECTIVE 2010/63/EU OF THE EUROPEAN PARLIAMENT AND OF THE COUNCILLe Scienze Web News
on the protection of animals used for scientific purposes.
http://www.lswn.it/comunicati/stampa/2012/sperimentazione_animale_recepire_direttiva_europea_per_garantire_rispetto_animali_e_attivita_ricerca
Archeologi della Sapienza scoprono a Malta un’agata con iscrizioni cuneiformi del II millennio a.C: si tratta del reperto più occidentale mai rinvenuto.
Leggi http://www.lswn.it/comunicati/stampa/2011/scoperta_a_malta_agata_con_iscrizioni_cuneiformi_secondo_millennio_ac
Electron Diffraction Using Transmission Electron MicroscopyLe Scienze Web News
Electron diffraction via the transmission electron microscope is a powerful method for characterizing the structure of materials, including perfect crystals and defect structures. The advantages of elec- tron diffraction over other methods, e.g., x-ray or neutron, arise from the extremely short wavelength (≈2 pm), the strong atomic scattering, and the ability to exam- ine tiny volumes of matter (≈10 nm3). The NIST Materials Science and Engineer- ing Laboratory has a history of discovery and characterization of new structures through electron diffraction, alone or in combination with other diffraction methods. This paper provides a survey of some of this work enabled through electron mi- croscopy.
Metallic Phase with Long-Range Orientational Order and No Translational Symmetry
D. Shechtman and I. Blech
Department of Materials Engineering, Israel Institute of Technology Technion, 3200 Haifa, Israel
and
D. Gratias
Centre d'Etudes de Chimie Metallurgique, Centre National de la Recherche Scientiftque, F-94400 Vitry, France
and
J. W. Cahn
Center for Materials Science, National Bureau ofStandards, Gaithersburg, Maryland 20760
(Received 9 October 1984)
We have observed a metallic solid (Al —14-at. /o-Mn) with long-range orientational order, but with icosahedral point group symmetry, which is inconsistent with lattice translations. Its diffraction spots are as sharp as those of crystals but cannot be indexed to any Bravais lattice. The solid is metastable and forms from the melt by a first-order transition.
12 novembre 2011 Cerimonia Ufficiale UNESCO "A come Acqua"Le Scienze Web News
Settimana di Educazione allo Sviluppo Sostenibile promossa dalla Commissione Nazionale Italiana per l’UNESCO “A come Acqua”.
Conferimento della Medaglia Spadolini
Premio Eco and the City Giovanni Spadol
12 novembre 2011 Palazzo Incontri, via de Pucci 1, Firenze Ore 15,00
Autore Prof. Roberto Bertagnolio
Articolo su LSWN.it:
"Programmi, equazioni, algoritmi, simmetria e visione"
http://www.lswn.it/miscellanea/articoli/programmi_equazioni_algoritmi_simmetria_e_visione
Mostra e convegno internazionale su energie rinnovabili e generazione distribuita
Per informazioni
http://www.lswn.it/eventi/esposizioni/2011/solar_expo_2011_mostra_convegno_internazionale_energie_rinnovabili_generazione_distribuita
Mobilità autonoma dei bambini: una necessità per loro, una risorsa per la cittàLe Scienze Web News
Convegno Unesco, Roma, 5 novembre 2010
Francesco Tonucci
Istituto di Scienze e Tecnologie della Cognizione Consiglio Nazionale delle Ricerche
Settimana Unesco di Educazione allo Sviluppo Sostenibile 2010
L'autonomia di movimento dei bambini: una necessità per loro, una risorsa per...Le Scienze Web News
Francesco Tonucci
Istituto di Scienze e Tecnologie della Cognizione (ISTC) Del Consiglio Nazionale delle Ricerche (CNR)
Responsabile del progetto internazionale “La città delle bambine e dei bambini”
Settimana Unesco di Educazione allo Sviluppo Sostenibile 2010
Centro di ricerca per il Trasporto e la Logistica
Francesco Filippi
Sapienza Università di Roma
Settimana Unesco di Educazione allo Sviluppo Sostenibile 2010
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists Saeid Safari
Preoperative Management of Patients on GLP-1 Receptor Agonists like Ozempic and Semiglutide
ASA GUIDELINE
NYSORA Guideline
2 Case Reports of Gastric Ultrasound
Anti ulcer drugs and their Advance pharmacology ||
Anti-ulcer drugs are medications used to prevent and treat ulcers in the stomach and upper part of the small intestine (duodenal ulcers). These ulcers are often caused by an imbalance between stomach acid and the mucosal lining, which protects the stomach lining.
||Scope: Overview of various classes of anti-ulcer drugs, their mechanisms of action, indications, side effects, and clinical considerations.
The prostate is an exocrine gland of the male mammalian reproductive system
It is a walnut-sized gland that forms part of the male reproductive system and is located in front of the rectum and just below the urinary bladder
Function is to store and secrete a clear, slightly alkaline fluid that constitutes 10-30% of the volume of the seminal fluid that along with the spermatozoa, constitutes semen
A healthy human prostate measures (4cm-vertical, by 3cm-horizontal, 2cm ant-post ).
It surrounds the urethra just below the urinary bladder. It has anterior, median, posterior and two lateral lobes
It’s work is regulated by androgens which are responsible for male sex characteristics
Generalised disease of the prostate due to hormonal derangement which leads to non malignant enlargement of the gland (increase in the number of epithelial cells and stromal tissue)to cause compression of the urethra leading to symptoms (LUTS
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...i3 Health
i3 Health is pleased to make the speaker slides from this activity available for use as a non-accredited self-study or teaching resource.
This slide deck presented by Dr. Kami Maddocks, Professor-Clinical in the Division of Hematology and
Associate Division Director for Ambulatory Operations
The Ohio State University Comprehensive Cancer Center, will provide insight into new directions in targeted therapeutic approaches for older adults with mantle cell lymphoma.
STATEMENT OF NEED
Mantle cell lymphoma (MCL) is a rare, aggressive B-cell non-Hodgkin lymphoma (NHL) accounting for 5% to 7% of all lymphomas. Its prognosis ranges from indolent disease that does not require treatment for years to very aggressive disease, which is associated with poor survival (Silkenstedt et al, 2021). Typically, MCL is diagnosed at advanced stage and in older patients who cannot tolerate intensive therapy (NCCN, 2022). Although recent advances have slightly increased remission rates, recurrence and relapse remain very common, leading to a median overall survival between 3 and 6 years (LLS, 2021). Though there are several effective options, progress is still needed towards establishing an accepted frontline approach for MCL (Castellino et al, 2022). Treatment selection and management of MCL are complicated by the heterogeneity of prognosis, advanced age and comorbidities of patients, and lack of an established standard approach for treatment, making it vital that clinicians be familiar with the latest research and advances in this area. In this activity chaired by Michael Wang, MD, Professor in the Department of Lymphoma & Myeloma at MD Anderson Cancer Center, expert faculty will discuss prognostic factors informing treatment, the promising results of recent trials in new therapeutic approaches, and the implications of treatment resistance in therapeutic selection for MCL.
Target Audience
Hematology/oncology fellows, attending faculty, and other health care professionals involved in the treatment of patients with mantle cell lymphoma (MCL).
Learning Objectives
1.) Identify clinical and biological prognostic factors that can guide treatment decision making for older adults with MCL
2.) Evaluate emerging data on targeted therapeutic approaches for treatment-naive and relapsed/refractory MCL and their applicability to older adults
3.) Assess mechanisms of resistance to targeted therapies for MCL and their implications for treatment selection
These lecture slides, by Dr Sidra Arshad, offer a quick overview of physiological basis of a normal electrocardiogram.
Learning objectives:
1. Define an electrocardiogram (ECG) and electrocardiography
2. Describe how dipoles generated by the heart produce the waveforms of the ECG
3. Describe the components of a normal electrocardiogram of a typical bipolar leads (limb II)
4. Differentiate between intervals and segments
5. Enlist some common indications for obtaining an ECG
Study Resources:
1. Chapter 11, Guyton and Hall Textbook of Medical Physiology, 14th edition
2. Chapter 9, Human Physiology - From Cells to Systems, Lauralee Sherwood, 9th edition
3. Chapter 29, Ganong’s Review of Medical Physiology, 26th edition
4. Electrocardiogram, StatPearls - https://www.ncbi.nlm.nih.gov/books/NBK549803/
5. ECG in Medical Practice by ABM Abdullah, 4th edition
6. ECG Basics, http://www.nataliescasebook.com/tag/e-c-g-basics
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...VarunMahajani
Disruption of blood supply to lung alveoli due to blockage of one or more pulmonary blood vessels is called as Pulmonary thromboembolism. In this presentation we will discuss its causes, types and its management in depth.
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?bkling
Are you curious about what’s new in cervical cancer research or unsure what the findings mean? Join Dr. Emily Ko, a gynecologic oncologist at Penn Medicine, to learn about the latest updates from the Society of Gynecologic Oncology (SGO) 2024 Annual Meeting on Women’s Cancer. Dr. Ko will discuss what the research presented at the conference means for you and answer your questions about the new developments.
Ethanol (CH3CH2OH), or beverage alcohol, is a two-carbon alcohol
that is rapidly distributed in the body and brain. Ethanol alters many
neurochemical systems and has rewarding and addictive properties. It
is the oldest recreational drug and likely contributes to more morbidity,
mortality, and public health costs than all illicit drugs combined. The
5th edition of the Diagnostic and Statistical Manual of Mental Disorders
(DSM-5) integrates alcohol abuse and alcohol dependence into a single
disorder called alcohol use disorder (AUD), with mild, moderate,
and severe subclassifications (American Psychiatric Association, 2013).
In the DSM-5, all types of substance abuse and dependence have been
combined into a single substance use disorder (SUD) on a continuum
from mild to severe. A diagnosis of AUD requires that at least two of
the 11 DSM-5 behaviors be present within a 12-month period (mild
AUD: 2–3 criteria; moderate AUD: 4–5 criteria; severe AUD: 6–11 criteria).
The four main behavioral effects of AUD are impaired control over
drinking, negative social consequences, risky use, and altered physiological
effects (tolerance, withdrawal). This chapter presents an overview
of the prevalence and harmful consequences of AUD in the U.S.,
the systemic nature of the disease, neurocircuitry and stages of AUD,
comorbidities, fetal alcohol spectrum disorders, genetic risk factors, and
pharmacotherapies for AUD.
ARTIFICIAL INTELLIGENCE IN HEALTHCARE.pdfAnujkumaranit
Artificial intelligence (AI) refers to the simulation of human intelligence processes by machines, especially computer systems. It encompasses tasks such as learning, reasoning, problem-solving, perception, and language understanding. AI technologies are revolutionizing various fields, from healthcare to finance, by enabling machines to perform tasks that typically require human intelligence.
Prix Galien International 2024 Forum ProgramLevi Shapiro
June 20, 2024, Prix Galien International and Jerusalem Ethics Forum in ROME. Detailed agenda including panels:
- ADVANCES IN CARDIOLOGY: A NEW PARADIGM IS COMING
- WOMEN’S HEALTH: FERTILITY PRESERVATION
- WHAT’S NEW IN THE TREATMENT OF INFECTIOUS,
ONCOLOGICAL AND INFLAMMATORY SKIN DISEASES?
- ARTIFICIAL INTELLIGENCE AND ETHICS
- GENE THERAPY
- BEYOND BORDERS: GLOBAL INITIATIVES FOR DEMOCRATIZING LIFE SCIENCE TECHNOLOGIES AND PROMOTING ACCESS TO HEALTHCARE
- ETHICAL CHALLENGES IN LIFE SCIENCES
- Prix Galien International Awards Ceremony
Lung Cancer: Artificial Intelligence, Synergetics, Complex System Analysis, S...Oleg Kshivets
RESULTS: Overall life span (LS) was 2252.1±1742.5 days and cumulative 5-year survival (5YS) reached 73.2%, 10 years – 64.8%, 20 years – 42.5%. 513 LCP lived more than 5 years (LS=3124.6±1525.6 days), 148 LCP – more than 10 years (LS=5054.4±1504.1 days).199 LCP died because of LC (LS=562.7±374.5 days). 5YS of LCP after bi/lobectomies was significantly superior in comparison with LCP after pneumonectomies (78.1% vs.63.7%, P=0.00001 by log-rank test). AT significantly improved 5YS (66.3% vs. 34.8%) (P=0.00000 by log-rank test) only for LCP with N1-2. Cox modeling displayed that 5YS of LCP significantly depended on: phase transition (PT) early-invasive LC in terms of synergetics, PT N0—N12, cell ratio factors (ratio between cancer cells- CC and blood cells subpopulations), G1-3, histology, glucose, AT, blood cell circuit, prothrombin index, heparin tolerance, recalcification time (P=0.000-0.038). Neural networks, genetic algorithm selection and bootstrap simulation revealed relationships between 5YS and PT early-invasive LC (rank=1), PT N0—N12 (rank=2), thrombocytes/CC (3), erythrocytes/CC (4), eosinophils/CC (5), healthy cells/CC (6), lymphocytes/CC (7), segmented neutrophils/CC (8), stick neutrophils/CC (9), monocytes/CC (10); leucocytes/CC (11). Correct prediction of 5YS was 100% by neural networks computing (area under ROC curve=1.0; error=0.0).
CONCLUSIONS: 5YS of LCP after radical procedures significantly depended on: 1) PT early-invasive cancer; 2) PT N0--N12; 3) cell ratio factors; 4) blood cell circuit; 5) biochemical factors; 6) hemostasis system; 7) AT; 8) LC characteristics; 9) LC cell dynamics; 10) surgery type: lobectomy/pneumonectomy; 11) anthropometric data. Optimal diagnosis and treatment strategies for LC are: 1) screening and early detection of LC; 2) availability of experienced thoracic surgeons because of complexity of radical procedures; 3) aggressive en block surgery and adequate lymph node dissection for completeness; 4) precise prediction; 5) adjuvant chemoimmunoradiotherapy for LCP with unfavorable prognosis.
Flu Vaccine Alert in Bangalore Karnatakaaddon Scans
As flu season approaches, health officials in Bangalore, Karnataka, are urging residents to get their flu vaccinations. The seasonal flu, while common, can lead to severe health complications, particularly for vulnerable populations such as young children, the elderly, and those with underlying health conditions.
Dr. Vidisha Kumari, a leading epidemiologist in Bangalore, emphasizes the importance of getting vaccinated. "The flu vaccine is our best defense against the influenza virus. It not only protects individuals but also helps prevent the spread of the virus in our communities," he says.
This year, the flu season is expected to coincide with a potential increase in other respiratory illnesses. The Karnataka Health Department has launched an awareness campaign highlighting the significance of flu vaccinations. They have set up multiple vaccination centers across Bangalore, making it convenient for residents to receive their shots.
To encourage widespread vaccination, the government is also collaborating with local schools, workplaces, and community centers to facilitate vaccination drives. Special attention is being given to ensuring that the vaccine is accessible to all, including marginalized communities who may have limited access to healthcare.
Residents are reminded that the flu vaccine is safe and effective. Common side effects are mild and may include soreness at the injection site, mild fever, or muscle aches. These side effects are generally short-lived and far less severe than the flu itself.
Healthcare providers are also stressing the importance of continuing COVID-19 precautions. Wearing masks, practicing good hand hygiene, and maintaining social distancing are still crucial, especially in crowded places.
Protect yourself and your loved ones by getting vaccinated. Together, we can help keep Bangalore healthy and safe this flu season. For more information on vaccination centers and schedules, residents can visit the Karnataka Health Department’s official website or follow their social media pages.
Stay informed, stay safe, and get your flu shot today!
2. Chemical Reviews Review
Figure 1. Recoverin 3D structure (left) and correspondent adjacency matrix.
95 burden required for a complete kinetic representation. expression correlation networks) for which no such support is 139
96 Mathematics and chemistry meet on the common ground of possible. 140
97 the chemical reaction network theory (CNRT) that is explicitly The inter-residue contact network has been yet largely 141
98 aimed at analyzing complex biochemical reaction networks in explored in terms of inter-residue contacts frequencies under 142
99 terms of their topological emerging features.10−13 the quasichemical approximation;18−20 as a matter of fact, in 143
100 Nowadays, many different fields of investigation ranging the seminal work of Miyazawa and Jernigan,18 the amino acid 144
101 from systems biology to electrical engineering, sociology, and hydrophobicity is assessed on the basis of the frequency of 145
102 statistical mechanics converge into the shared operational contacts of the corresponding residues as emerging from the 146
103 paradigm of complex network analysis.14 A massive advance- analysis of a large number of structures. 147
104 ment in the elucidation of general behavior of network systems In this way, residues involved more frequently in noncovalent 148
105 made possible the generation of brand new graph theoretical interactions (mainly of hydrophobic nature, for hypothesis) are 149
106 descriptors, at both single node and entire graph level, that addressed to be of similar hydrophobic character. Application 150
107 could be useful in many fields of chemistry. and confirmation of this view emerge from more recent 151
108 More specifically, in this review we will deal with the protein works,19,21 where the thermal stability of proteins belonging to 152
109 3D structures in terms of contact networks between amino acid thermophiles or psychrophiles has been inspected through the 153
110 residues. This case allows for a straightforward formalization in inter-residue interaction potential. The main result is that a 154
111 topological terms: the role of nodes (residues) and edges characteristic distribution of inter-residues is able to provide the 155
112 (contacts) is devoid of any ambiguity and the introduction of protein structure with the required flexibility to adapt to the 156
113 van der Waals radii of amino acids allows us to assign a environment. 157
114 motivated threshold for assigning contacts and building the The two above referenced works19,21 are, in any case, only a 158
115 network.15−17
statistical application over a huge number of proteins; what we 159
116 On the other hand, the Protein Data Bank (PDB) collects
really want to know is the character of information about a 160
117 thousands of very reliable X-ray-resolved molecular structures,
single and specific molecule that we can derive from its residue 161
118 allowing scientists to perform sufficiently populated statistical
contact graph. 162
119 enquiries to highlight relevant shared properties of protein
A very immediate example of this single molecule 163
120 structures or to go in-depth into specific themes (e.g.,
121 topological signatures of allostery), as well as to identify information is the fact that protein secondary structure can 164
122 residues potentially crucial for activity and stability of proteins. be reproduced with no errors on the sole basis of an adjacency 165
123 From a purely theoretical point of view, the reduction of a matrix.22 Similar considerations hold true for protein folding 166
124 protein structure (that in its full rank corresponds to the three- rate,23−26 while normal-mode analysis confirmed that mean 167
125 dimensional coordinates relative to all the atoms of the square displacement of highly contacted residues is substantially 168
126 molecule) to a binary contact matrix between the α-carbons of limited (nearly 20% of maximal movement range27). From 169
127 the residues represents a dramatic collapse. How many relevant another perspective, the presence of highly invariant patterns of 170
128 properties of protein 3D structures (and consequently of graph descriptors shared by all the proteins, irrespective of their 171
129 possible consequences in terms of protein physiological role) general shape and size, points to still unknown mesoscopic 172
130 are kept alive (and, hopefully, exalted by the filtering out of not invariants (formally an analogue to valence considerations) on 173
131 relevant information) by the consideration of a protein as a the very basis of protein-like behavior, irrespective for both 174
132 contact network? How firmly based is the guess that adjacency fibrous and globular structures.28,29 The scope of this review is, 175
133 matrices having as rows and columns amino acid residues (see by briefly discussing some applications in this rapidly emerging 176
f1 134 Figure 1) could in the future play the same role the structural field, to sketch an at least initial answer to the quest for a new 177
135 formula plays for organic chemistry? Relying on a single “structural formula” language for proteins. This quest will be 178
136 nonambiguous and physically motivated ordering of nodes (the pursued in the following chapters by presenting side-by-side the 179
137 primary structure) dramatically enlarges the realism of contact different complex network invariants developed by graph 180
138 networks with respect to other kinds of networks (e.g., gene theory and their protein counterparts. 181
B dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
3. Chemical Reviews Review
2. GRAPH THEORY AND PROTEIN CONTACT ⎧ Ai , j = w(vi , vj) if (vi , vj) ∈ E
⎪
182 NETWORKS Aij = ⎨
⎪
⎩ Ai , j = 0 otherwise
2.1. Elements of Graph Theory
183 The classic Könisberg bridge problem introduced graph theory The degree of a node in a weighted graph is defined as 220
184 in 18th century. The problem had the following formulation:
N
185 does there exist a walk crossing each of the seven bridges of
186 Könisberg exactly once? The solution to this problem appeared k(v) = ∑ w(ui , v)
187 in “Solutio Problematis ad geometriam situs pertinentis” in i=1
188 1736 by Euler.30 This was the first time a problem was codified
where ui ∈ N(v). 221
189 in terms of nodes and edges linking nodes. This structure was
190 called a graph. 2.2. Protein Contact Networks (PCNs)
191 A graph G is a mathematical object used to model complex A protein structure is a complex three-dimensional object, 222
192 structures and it is made of a finite set of vertices (or nodes) V formally defined by the coordinates in 3D space of its 223
193 and a collection of edges E connecting two vertices. atoms.31,32 Since the first works on the subject in the early 224
194 A graph G = (V, E) can be represented as a plane figure by 1960s,33 a large number of protein molecular structures has 225
195 drawing a line between two nodes u and v and an edge e = (u, been resolved, now accessible on devoted web databases.34 The 226
f2 196 v) ∈ E (Figure 2). large availability of protein molecular structures has not solved 227
yet many of the issues regarding the strict relationship between 228
structure and function in the protein universe. 229
Thus, an emerging need in protein science is to define simple 230
descriptors, able to describe each protein structure with few 231
numerical variables, hopefully representative of the functionally 232
Figure 2. Example of an undirect graph comprising two nodes and an relevant properties of the analyzed structure. 233
edge. Protein structure and function rely on the complex network 234
of inter-residue interactions that intervene in forming and 235
keeping the molecular structure and in the protein biological 236
activity. 237
197 A graph G = (V, E) can be represented by its adjacency Thus, the residues interactions are a good starting point to 238
198 matrix A; given an order of V = {v1, v2, ... vn}, we define the define the protein interaction network;20,27,35 in this frame- 239
199 generic element of the matrix Ai,j as follows: work, the molecular structure needs to be translated into a 240
simpler picture, cutting out the redundant information 241
⎧ Ai , j = 1 if (vi , vj) ∈ E
⎪
embedded in the complete spatial position of all atoms. 242
Aij = ⎨ The most immediate choice is collapsing it into its α-carbon 243
⎪
⎩ Ai , j = 0 otherwise location (thereinafter indicated as Cα): correspondingly, the 244
position of the entire amino acid in the sequence is collapsed 245
200 The adjacency matrix of a graph is unique with respect to the into the corresponding Cα. 246
201 chosen ordering of nodes. In the case of proteins, where the The spatial position of Cα is still reminiscent of the protein 247
202 ordering of nodes (residues) corresponds to the residue backbone; thus residues that are immediately close in sequence 248
203 sequence (primary structure), we can state that its correspond- are separated by a length of 3−4 Å, corresponding to the 249
204 ing network is unique. This is one extremely strong peptide bond length36 (see Figure 3); other α-carbons have a 250 f3
205 consequence that establishes a 1 to 1 correspondence between position that recalls the secondary domains and still reproduce, 251
206 the molecule and its corresponding graph. even in a very bare representation, the key features of the three- 252
207 Let v ∈ V be a vertex of a graph G; the neighborhood of v is dimensional structure. 253
208 the set N(v) = {u ∈ G | e(u,v) ∈ E}. Two vertices u and v are As soon as the complex protein structure architecture has 254
209 adjacent or neighbors, when e = (u,v) ∈ E (u ∈ N(v) or v ∈ been reduced to a simpler picture in terms of Cα position, the 255
210 N(u)). The degree ki of the ith node is the number of its spatial topology can be further reduced to a contact topology 256
211 neighbors, defined on the basis of the adjacency matrix as that represents the network of inter-residue interactions, 257
primarily responsible for the protein’s three-dimensional 258
N
structure and activity. Thus, the interaction topology is derived 259
ki = ∑ Aij by the spatial distribution of residues in the crystal three- 260
j=1 dimensional structure and represents the overall intramolecular 261
potential. 262
212 When ki = 0, the ith node is said to be isolated in G, whereas if Specifically, starting from the Cα spatial distribution, the 263
213 ki = 1, it is said to be a leaf of the graph. distance matrix d = {dij} is computed, the generic element dij 264
214 Information may be attached to edges, in this case we call the being the Euclidean distance in the 3D space between the ith 265
215 graph weighted and we refer to the weights as “costs”. A and jth residues (holding the sequence order). The interaction 266
216 weighted graph is defined as G = (V, E, W), where W is a topology is then computed on the basis of d: if the distance dij 267
217 function assigning to each edge of the graph a weight: falls into a given spatial interval (said cutoff), a link exists 268
W: E → between the ith and the jth residues. The definition of the type 269
of the graph (unweighted or weighted) is made in order to 270
218 The adjacency matrix A of a weighted graph is defined as describe a given kind of interaction, in a more or less detailed 271
219 follows: fashion. 272
C dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
4. Chemical Reviews Review
represented by all of its atoms. Then, the distance matrix is 291
computed over all protein atoms, which are labeled according 292
to the residue they belong to. The strength of the interaction 293
between two residues is measured as the number of their atoms 294
whose distance lies within .15,39,46−49 295
Eventually, a straightforward way to establish a weighted 296
protein contact network is to take the inverse of the distance 297
among two residues as a direct measurement of their mutual 298
interaction: the closer they lie, the stronger their mutual 299
interaction.50,51 300
Another kind of representation is based on the same criterion 301
but adopts as nodes the 20 different amino acids, which are 302
combined through the peptide bond backbone in the protein 303
primary structure. The link between two residues is represented 304
by the number of links the residues of those types establish in 305
the three-dimensional structure, according to the distance 306
matrix d and the cutoff interval , as a rule. This method can be 307
applied on an ensemble of protein structures,43,52 in order to 308
find a common rule of protein structure construction, in terms 309
of more probable contacts between residues. 310
This representation, while keeping track of the nature of the 311
interacting residues, destroys the one-to-one correspondence 312
Figure 3. Geometry of the peptide bond: the upper threshold of 8 Å,
with the original 3D structure, given that different structures 313
commonly introduce in the of PCNs, roughly corresponds to two
peptide bond lengths.36 can give rise to the same representation in a way analogously to 314
the structure isomerism in organic chemistry. Figure 4 reports 315 f4
273 The choice of determines the kind of interactions included the two kind of formulas. 316
274 in the analysis.17,37 Most authors15,16,38,39 consider only an The first emerging property of the PCNs is the degree of the 317
275 upper threshold (around 8 Å) to cut off negligible interactions; corresponding graph, i.e., the average number of links each 318
276 some others, conversely, introduce also a lower limit, around 4 node (residue) establishes with neighbors. It is a direct measure 319
277 Å, that corresponds to the average value of the peptide bond of connectivity attitude of residues within the interaction 320
278 length, so to eliminate the “noise” due to the “obliged” contacts network and it is strictly linked to the attitude of residue to 321
279 coming from sequence proximity. In this way, only significant establish noncovalent interactions with other resi- 322
280 noncovalent interactions are included in the analysis, with the dues.28,38,40,46,49,54−56 The average degree, on the other hand, 323
281 purpose of including only those interactions that may be is a measure of the overall protein connectivity that is a rough 324
282 modified upon slight environment changes, such in the case of index of the protein stability. 325
283 biological response to environment stimuli. The contact density of a protein decreases exponentially with 326
284 Many authors use unweighted graphs to represent the number of residues; thus, bigger proteins are much less 327
285 PCNs,16,35,40−46 in order to infer several properties while compact than smaller ones, giving rise to bigger cavities and a 328
286 keeping minimal information. On the other hand, some other more fuzzy distinction between internal and external 329
287 groups propose a description for the PCNs using weighted milieu.57,58 330
288 graphs that is based on a side chain level reduction of the whole The degree distribution defines the graph model, allowing us 331
289 protein structure. In this case, all the information regarding the to classify the network into already established network classes 332
290 spatial position of atoms is kept and the single residue is endowed with specific features (e.g., random graphs, scale-free 333
Figure 4. Graph protein formulas: (a) contact map53 and (b) wheel diagram.27
D dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
5. Chemical Reviews Review
334 networks, regular lattices) as we will show in the next 3 × (the number of triangles on a graph)
335 paragraphs. C=
the number of connected triples of vertices (2.2) 391
2.3. Shortest Paths, Average Path Length, and Diameter
336 In a graph G, the distance spv,u between any two vertices v,u ∈ where a “triangle” corresponds to three vertices that are each 392
337 V is given by the length of the shortest path between the connected to each other and a “connected triple” means a 393
338 vertices, that is, the minimal number of edges that need to be vertex that is connected to an (unordered) pair of other 394
339 crossed to travel from vertex v to vertex u. The shortest path vertices. The factor of 3 in the numerator accounts for the fact 395
340 between two vertices is not necessarily unique, since different that each triangle contributes to three connected triples of 396
341 paths may exist with identical length. In a graph, if no path vertices, one for each of its three vertices; thus, the value of C 397
342 exists connecting two nodes v, u ∈ V, we say that those nodes lies strictly in the range from zero to one. 398
343 belong to different connected components; in such a case, we With regard to PCNs, the clustering coefficient referred to 399
344 call the graph disconnected. the ith residue measures the triangles number insisting on 400
345 All PCNs are connected graphs at first glance. The so-called it;15,28,29,38,40,45−47,49,56,75 thus, high clustering coefficient nodes 401
346 “percolation threshold” of a PCN can be estimated as the are central in communities with a large number of 402
347 number of edges to be destroyed in order for the PCN to lose interconnecting links, corresponding to high local stability. In 403
348 its connectivity. other words, we can infer that mutation producing depletion of 404
349 This concept becomes relevant when we focus on long-range such nodes may cause dramatic changes in the protein 405
350 contacts (i.e., contacts between residues far away on the structure.29 406
351 sequence,23,59 which were demonstrated to be of crucial 2.4.1. Spectral Clustering. The spectral analysis of a graph 407
352 importance in protein folding rates,23,25 as we will see in allows one to identify clusters in the network by minimizing the 408
353 more detail below. value of parameter Z defined as45 409
354 The diameter diam(G) = max{spv,u|v, u ∈ VG} of a graph is n n
defined as the maximal distance of any pair of vertices. The 1
355
average or characteristic length l(G) = ⟨spv,u⟩ is defined as the
Z= ∑ ∑ (xi − xj)2 Aij
356 2 i=1 j=1
357 average distance between all pairs of vertices; the average
358 inverse path length (efficiency) is defined as eff(G) = ⟨1/spv,u⟩; where xi and xj represent the position of nodes i and j in the 410
359 this descriptor is particularly suitable when components are network and Aij is the adjacency matrix. The minimum of Z 411
360 disconnected (in this case, the contribution of infinite distances corresponds to the second smallest eigenvalue of the Laplacian 412
361 corresponds to zero efficiency). matrix L of {Aij}, also known as the Kirchoff matrix, defined as 413
362 The shortest path spv,u between two residues of a PCNs
363 represents a molecular shortcut that connect the residues L=D−A
364 through a mutual interaction pathway. In this sense, the smaller
365 the spu,v, the tighter the relationship between the two nodes, where D is the degree matrix, which is a diagonal matrix in 414
366 which are strictly correlated, regardless of their distance in a which {Dii} = ki. Once L eigenvalues λ are computed, the 415
367 sequence. These tight relations are thought to be responsible second smallest eigenvalue λ2 corresponds to the minimum 416
368 for the allosteric response in protein ligand binding42,60−65 and value of Z (the first one provides a trivial solution45). The 417
369 in the concerted motions of distinct protein regions in protein components of the corresponding eigenvector v2, known as the 418
370 dynamics.64,66−71 Fiedler eigenvector, refer to single nodes and define two 419
371 In general, still preliminary evidence from our group (work clusters depending on the sign of each component. Nodes are 420
372 in progress) points to the average shortest path as the most parted into two clusters according to the sign of the 421
373 crucial network invariant to link topology to both molecules’ corresponding component in v2. This process can be iterated 422
374 dynamics and the general thermodynamical properties of the on both subnetworks until all the components of v2 show the 423
375 protein molecules. same sign. 424
2.4. Clustering on Graphs Identification of clusters in PCNs has a strong impact on 425
detecting structural and functional domains in pro- 426
376 Identifying clusters on a network is a more complicated task teins.50,54,76,77 The presence of folding clusters is a key point 427
377 than computing the average shortest path. The clustering in the molecular development of the funnel folding pathways 428
378 coefficient measures the cliquishness of a typical neighborhood theory, which provides the most reasonable molecular 429
379 (a local property). One possible definition is the follow- mechanism for protein folding, out of the random approaches 430
380 ing:29,30,40,72 let us define the clustering coefficient of the ith of residues, in order to form the favorable inter-residue 431
381 node Ci as interaction network, providing stability to the tertiary 432
the number of connected neighbor pairs structure.78,79 433
Ci = 1 The reliable identification of clusters in the PCNs allows for 434
k (k
2 i i
− 1) (2.1) the definition of descriptors at the single residue level, relying 435
382
on the PCNs partition structure. 436
383 where ki is the degree of the ith vertex; the average clustering 2.4.2. Intracluster and Extracluster Parameters. Once 437
384 coefficient C of the graph is the average of Ci values over all the clustering process is performed, two parameters, zi and Pi, 438
385 nodes. representing the modularity rate for each node,80 can be 439
386 For social networks, Ci and C have intuitive meanings: Ci computed. These two parameters are defined as 440
387 reflects the extent to which friends of i are also friends of each
388 other; thus, C measures the cliquishness of a typical friendship kis − kis
̅
389 circle. zi =
390 Another definition for C is73,74 σis (2.3) 441
E dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
6. Chemical Reviews Review
nM⎛ kis ⎞2
Pi = 1 − ∑ ⎜ ⎟
s=1 ⎝ i ⎠
k (2.4)
442
443 where kis is the number of links the ith node establishes with
444 nodes belonging to its own cluster si; k̅si is the average degree
445 for nodes in cluster si, σsi is the corresponding standard
446 deviation, and nM is the number of clusters to which the ith
447 node belongs to. The spectral clustering performs a “crispy”
448 partition, namely, clusters are disjoint sets of nodes, thus eq 2.4
449 becomes
⎛ kis ⎞2
Pi = 1 − ⎜ ⎟
450
⎝ ki ⎠ (2.5)
451 These parameters have been introduced to discriminate
452 nodes according to their topological role in the so-called
453 Guimerà and Amaral’s cartography,80 the aim of which is the
454 classification of nodes in a modular network, relying on intra-
455 and intermodule connectivities.81
456 In their seminal work,80 Guimerà and Amaral demonstrated
457 that the relative importance of each node in maintaining the
458 global graph connectivity can be traced back to its location in
459 the P, z plane.
460 Once the network is partitioned into a set of meaningful
461 communities, it is possible to compute statistics for how
462 connected each hub (a hub is a node having an extremely high
463 degree of connectivity) is both within its own community and
464 to other communities: hubs endowed with strong connections
465 within functional modules were assumed to be interacting with
466 their partners at once (party hubs); conversely, those with a
467 low correlation were assumed to link together multiple modules
468 (date hubs), playing a global role in the network. It is worth
469 stressing that although both hub types have similar essentiality
470 in the network, as the characteristic path length increases, Figure 5. P vs z plot (dentist’s chair) (a) for a single protein, where
471 deleting given hubs, the network begins to disintegrate, since each point identifies a residue, and (b) the superposition of structures
472 hubs provide the coordination between functional modules. To analysis of 1420 proteins58.
473 make a comparison, party hubs should correspond to Guimerà
474 “provincial hubs”, which have many links within their module
The strong invariance of the P, z portraits of PCNs 499
475 but few outside, whereas date hubs could be “nonhub
irrespective of both protein general shape and size is extremely 500
476 connectors” or “connector hubs”, both of which have links to
intriguing, given it suggests the existence of still hidden 501
477 several different modules; they could also fall into the “kinless”
mesoscopic principles of protein structures analogous to 502
478 roles, since very few nodes are actually found in these
valence rules in general chemistry. 503
479 categories.82 Considering network motifs, it was observed
480 that party hub network motifs control a local topological 2.5. Network Centralities
481 structure and stay together inside protein complexes, at a lower The centrality of a node deals with its topological features in 504
482 level of the network. On the other hand, date hub network network wiring. The term “central” stems from the origin of 505
483 motifs control the global topological structure and act as the this concept in the definition of key, central indeed, nodes of 506
484 connectors among signal pathways, at a high level of the social networks: people, in other words, that are responsible for 507
485 network. Network motifs should not be merely considered as a the stability and activity of the network. This “social science” 508
486 connection pattern derived from topological structures but also origin of the concept of centrality was found to have a 509
487 as functional elements organizing the modules for biological correlation in PCNs by Csermely.83 510
488 processes.67 Centrality can be computed in different ways, using different 511
489 Spectral clustering of PCNs produces characteristic P−z weights to evaluate and compare the importance of a node 512
490 diagrams, referred to as “dentist’s chair”, due to their (degree, clustering coefficient, for instance). They are almost 513
491 shape.28,29,58 This shape is strongly invariant with respect to equivalent definitions that point to the same attitude of central 514
f5 492 the protein molecule, as shown in Figure 5; panel a refers to a nodes, to establish strong local interactions, in their own local 515
493 typical diagram derived from the analysis of a single protein community, able to stabilize the whole network structure. 516
494 structure, while panel b shows the superposition of a structure The role of central nodes in modifying the network structure 517
495 analyses of 1420 proteins. The fact the general shape of the according to their centrality values is the starting point to define 518
496 graph remains substantially invariant on going from one to the property of centrality−lethality,81,84 which emerges as a key 519
497 1420 proteins is an impressive proof of the robustness of the P, element in the analysis of biological networks, where central 520
498 z organization of PCNs. nodes represent a prerequisite for the organism survival: for 521
F dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
7. Chemical Reviews Review
522 instance, a shortage or a depletion of a central protein in a σv , u(s)
523 protein−protein interaction network does lead to the death of betw(s) = ∑ ∑
σv , u
524 the organism.85 v∈V ,v≠s u∈V ,u≠s
525 Central nodes in PCNs correspond to residues crucial for
In biological networks (e.g., protein−protein interaction 580
526 both the protein structure folding and stability. Thus, the
network), the nodes with higher between-ness were demon- 581
527 centrality of a node can be a measure of the biological strated to be the main regulators.84,88,89 582
528 consequences of its mutation; for instance, the highly In PCNs, since the between-ness centrality is based on 583
529 detrimental mutation of hemoglobin that causes sickle cell shortest paths, it comes immediately clear that this index is 584
530 anemia is due to just a substitution of one residue (glutamic strongly linked to the centrality of nodes (residues) in terms of 585
531 acid is replaced in position 6 by valine) that produces dramatic their capability to transfer signals throughout the protein 586
532 changes in the protein structure and function. This effect is molecule.43,56,72,86 Thus, the depletion of residues having high 587
533 widely reflected in the high centrality value of this specific between-ness centrality values is supposed to interrupt the 588
534 residue.29 allosteric communication among regions of the proteins that lie 589
535 The easiest and the most natural way to define centrality is far apart. 590
536 provided by the so-called degree centrality that simply counts
2.6. Network Assortativity and Nodes Property
537 the number of connections for each node, its degree, i.e., the
Distribution 591
538 number of nodes it is directly connected with; in this case, the 90
539 degree centrality of a node vi corresponds to its degree. Hubs Newman suggested that an important driving factor in the 592
540 are, thus, the central nodes of a network, according to this formation of communities was the preference of nodes to 593
541 paradigm. connect to other nodes that possess similar characteristics; he 594
542 2.5.1. Path-Based Centralities: Closeness and Be- defined this behavior as assortativity. The concept of 595
543 tween-ness. Closeness centrality, as well as between-ness, assortativity is a very general one, so in the case of protein 596
544 belong to the class of shortest path-based centrality measures. structures, we could identify the “behavior” of different residues 597
545 The closeness centrality provides information about how close in terms of their hydrophobic/hydrophilic character so that an 598
546 a node is to all other nodes. The closeness of a node vi is assortative structure will correspond to a network in which 599
547 defined as similar hydrophobicity residues will be preferentially in contact 600
with each other compared to what is expected by pure chance. 601
1 In this example, the “behavior” of the nodes corresponds to a 602
c(vi) = n feature (hydrophobicity) independent of the pure network 603
∑ j = 1 spi , j
wiring and can be equated to a “coloring” of the nodes, whose 604
relations with the underlying topological support constituted by 605
548 The closeness centrality is connected to the aptitude of a network wiring is investigated. Along similar lines, we could 606
549 node to participate in the signal transmission throughout the think of assortative social networks in which friends (nodes in 607
550 protein structure. High closeness centrality nodes were direct contact) tends to share the same political ideas, income 608
551 demonstrated to correspond to residues located in the active classes, or professional activities. On a different heading, we can 609
552 site of ligand-binding proteins or to evolutionary conserved think of assortativity as an “internal” description of network 610
553 residues.41,52,72,86,87 wiring in which nodes are defined in terms of their connection 611
554 It is worth noting that closeness centrality, at odds with patterns. Actually, in some networks high-degree nodes 612
555 degree centrality, it is not solely based on local features of the preferentially connect to other high-degree nodes (assortative 613
556 network but takes into account the location of the node in the networks), whereas in other types of networks high-degree 614
557 global context of the network it is embedded into. In this nodes connect to low-degree nodes (disassortative networks); 615
558 respect, closeness, as well as between-ness, are genuine systemic in particular, numerical evidence from experimental data have 616
559 properties that are computed at the single node level, thus shown that many biological networks exhibit a negative 617
560 establishing a “top-down” causative process. This is probably assortativity coefficient and are therefore claimed to be 618
561 the reason for the efficiency of this kind of network invariants examples of disassortative mixing.40,91 619
562 to single out relevant general properties of the proteins. Assortativity r is defined as the Pearson correlation 620
563 This is formally analogous to what happens in basic coefficient of degrees at either ends of an edge, and it varies 621
564 chemistry, where the properties (i.e., acidity, electronegativity) as −1 ≤ r ≤ 1;92 r is a very simple measure of the probability of 622
565 of the hydrogen atom in the CH4 molecule are different from a high-degree node to form edges with other high-degree 623
566 those of the hydrogen atoms in H2O or H2 molecules, because nodes. When the r value is close to 1, the network is addressed 624
567 of the general molecular context they are embedded into. to as assortative, whereas values of r close to −1 are 625
568 This is the same philosophy of single node (residue) characteristic of disassortative networks. Random graphs are 626
569 descriptors, implicitly taking into account the whole context purely nonassortative networks, since by definition, links 627
570 and so overcoming a purely reductionist view. between nodes, in this case, 628
571 Between-ness measures the ability of a vertex to monitor N
572 communication between other vertices; every vertex that is part
573 of a shortest path between two other vertices can monitor and
k(v) = ∑ w(ui , v)
i=1
574 influence communication between them. In this view, a vertex
575 is central if lots of shortest paths connecting any two other are placed at random. 629
576 nodes cross it. Let σv,u denote the number of shortest paths In the case of external “coloring” assortativity, the index r, 630
577 between two vertices v, u ∈ V and let σv,u(s), where s ∈ V, be instead of being computed on the nodes degree, can be 631
578 the number of shortest paths between v and u crossing s; computed over the feature of interest, the one used to “color” 632
579 trivially σv,u ≥ σv,u(s). the nodes. 633
G dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
8. Chemical Reviews Review
634 Thus, in a recent work,57 Di Paola et al. demonstrated the One is called Gn,m and is the set of all graphs consisting of n 687
635 lack of any clearly defined “hydrophobic core” in proteins, for vertices and m edges, and it is built by throwing down m edges 688
636 which the arrangement of fractal structures was demonstrated between vertex pairs chosen at random from n initially 689
637 not to have a clear-cut separation between internal and external unconnected vertices. 690
638 milieu by means of network assortativity measures based on the The other is called Gn,p and it is the set of all graphs 691
639 hydrophobicity of nodes. Moreover, the presence of both consisting of n vertices, where each pair is connected together 692
640 assortative and disassortative structuring (hydrophobic−hydro- with independent probability p. In order to generate a graph 693
641 phobic and hydrophilic−hydrophobic) in proteins highlighted sampled uniformly at random from the set Gn,p, initially 694
642 the presence of different “folding logic” contemporarily present unconnected vertices are taken and each pair of them is joined 695
643 in the protein world, probably as a consequence of the varying with an edge with probability p (1 − p being the probability of 696
644 relevance of hydrophobic and electronic forces in the folding being unconnected). Thus, the presence or absence of an edge 697
645 process. between two vertices is independent of the presence or absence 698
646 Generally speaking, the distribution of a given feature of the of any other edge, so that each edge may be considered to be 699
647 nodes can be explored through the combined definition of present with independent probability p. The two models are 700
648 diadicity and heterophilicity,93 measuring the tendency of essentially equivalent in the limit of a large number of nodes n. 701
649 nodes with similar properties to form links. Given a key Since Gn,p is somewhat simpler to work with than Gn,m, it is 702
650 physical property, if nodes show an attitude to establish usual to refer to it as a random graph Gn,p. 703
651 preferentially links with similar nodes, the network is named as A vertex in a random graph is connected with equal 704
652 dyadic, otherwise it is said to be antidyadic or heterophilic.93 probability p to each of the N − 1 other vertices in the graph, 705
653 Let n1 and n0 respectively denote the number of node and hence, the probability pk that it has degree k is given by the 706
654 possessing or not a specific property; e10 and e11 are the number binomial distribution 707
655 of edges connecting homologous and heterelogous nodes,
⎛N ⎞
656 respectively. The heterophilicity score H is then defined as pk = ⎜ ⎟pk (1 − p)N − k
e ⎝k⎠ (2.10) 708
H = 10
e10,r (2.6) Noting that the average degree of a vertex in the network is z = 709
657
(N − 1)p, we can also write this as 710
658 where e10,r is the random value in case of uniform distribution
659 of the property among nodes that depends on the number of (N − 1)! zk ⎛ z ⎞
N−k
pk = ⎜1 − ⎟
660 possible edges E = N(N − 1)/2, N = n1 + n0 beingthe number k! (N − 1 − k)! (N − 1)k ⎝ N − 1⎠
661 of nodes:
z k e −z
e10,r = En1(N − n1) ≃
662 (2.7) k! (2.11) 711
663 Analogously, as for the homologous contacts, it is defined the where the second equality gets exact as N → ∞; in this case, pk 712
664 dyadicity D as corresponds to the bell-shaped curve that peaks on the average 713
e value (Figure 6b). 714 f6
D = 11
665
e11,r (2.8)
666 and the corresponding value for random homologous nodes is
n1(n1 − 1)
e11,r = E
667 2 (2.9)
668 Thus, dyadic networks have D values larger than 1 and, on
669 the other hand, H values lower than unity.
670 The above-described network invariants provide a descrip-
671 tion that can be traced back to the single node of a network, but Figure 6. Random graph: (a) a sample picture, where most nodes have
672 the effective values of the descriptors strongly depend on the three or four links, and (b) the bell-shaped degree distribution.
673 general wiring architecture of the whole graph, again a systemic
674 top-down causation metric. The dyadic character of PCNs was
675 exploited by Alves and colleagues94 to define simple hydro- Random graphs have been employed extensively as models 715
676 phobicity scores to profile protein structure. Single residue of real-world networks of various types, particularly in 716
677 hydrophobicity was demonstrated to be strongly correlated epidemiology,74 where the spreading of a disease through a 717
678 with the corresponding network invariants:56 these systemic community strongly depends on the pattern of contacts 718
679 properties strictly depend upon the “general class” the specific between infected subjects and those susceptible to it. 719
680 graph pertains to. Below we will briefly present the main classes However, as a model of a real-world network, a random 720
681 of wiring architectures. graph has some serious shortcomings. Perhaps the most serious 721
one is its degree distribution, which is quite unlike those seen in 722
2.7. Models of Graphs most real-world networks.92 On the other hand, the random 723
682 2.7.1. Random Graphs. One of the simplest and oldest graph has many desirable properties; specifically, many of its 724
683 network models is the random graph model,95 which was properties can be calculated exactly.92 725
684 introduced by Solomonoff and Rapoport96 and studied The random graph model has been applied to PCNs to test 726
685 extensively by Erdös and Rènyi;97−99 according to their their connectivity (degree) distribution.48,75 Specifically, the 727
686 works, there are two different random graph models. protein dynamic properties have been explored in terms of 728
H dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
9. Chemical Reviews Review
729 random graphs, since the unbiased corresponding network all nodes, and for γ = 2, a hub network emerges, with the largest 772
730 dynamics can be put into the perspective of the random hubs being in contact with a large fraction of all nodes.107 In 773
731 evolution of the protein structure, due to random, Brownian general, the unusual properties of scale-free networks are valid 774
732 motion of protein segments to get up to the final, stable only for γ < 3, such as a high degree of robustness against 775
733 conformation.100 accidental node failures.85 For γ > 3, however, most unusual 776
734 Further, the generic random graph model is introduced as a features are absent, and in many respects, the scale-free network 777
735 reference to test the property of the network as a specialized behaves like a random one.85 As for the World Wide Web, 778
736 random graph (small-world net- Barabàsi105 found that the value of γ for incoming links was 779
737 work).15,35,38,40,41,43,47,49,72,101−103 This comparison has a tight approximately 2; this means that any node has roughly a 780
738 link with the common assumption of the random coil structure probability 4 times bigger to have half the number of incoming 781
739 as being a reference state in folding thermodynamics: the links than another node. 782
740 random coil has the corresponding translation in terms of a Different from a Poisson degree distribution of random 783
741 “graph formula” into the random graph model that represents a networks, a power law distribution does not have a peak, but it 784
742 random network of residue interactions, corresponding to a is described by a continuously decreasing function (Figure 8): 785 f8
743 random distribution of the inter-residue distance.
744 In their work,104 Bartoli and colleagues demonstrated PCNs
745 are very far from random graph behavior, this was particularly
746 evident when they projected simulated networks together with
747 real PCNs in the bidimensional space, spanned by the
748 clustering coefficient and characteristic path length (see Figure
f7 749 7).
Figure 8. Scale-free networks: (a) a sample scale-free networks, in
which few nodes have many links, and (b) the degree distribution of
the scale-free graph power law.106
in this case, it is evident that a specific characteristic average 786
degree does not exist; in other words, these networks do not 787
converge toward a characteristic degree, at increasing number 788
Figure 7. Characteristic path length vs Clustering Coefficient (Figure of nodes. On the contrary, in scale-free networks, the average 789
3 in ref 104): sample protein classes are labeled as CA#, the label degree progressively increases with sampling dimension, 790
“random” refers to collection of random graphs, whereas “regular”
points to periodic lattices.
because the (very rare) high-degree nodes are sampled with a 791
higher probability. The lack of a characteristic degree is on the 792
basis of the denomination “scale free” for this kind of 793
750 The authors demonstrated the difference between random architecture. 794
751 graph and contact maps derive from the existence of the This is in strong contrast to random networks, for which the 795
752 covalent backbone, that imposes very strict constraints to the degree of all nodes is in the vicinity of the average degree, 796
753 contact that can be established between residues. This feature which could be considered typical. However, as Barabàsi and 797
754 makes PCNs to more similar to the so-called scale-free graphs. colleagues wrote in,107 scale-free networks could easily be called 798
755 2.7.2. Scale-Free Graphs. Since many years from the scale-rich as well, as their main feature is the coexistence of 799
756 seminal work of Erdös and Rènyi,97 all complex networks are nodes of widely different degrees (scales), ranging from nodes 800
757 treated commonly as random graphs. This paradigm was with one or two links to major hubs. 801
758 outdated by the pioneristic work of Barabàsi,105 in which the In contrast to the democratic distribution of links typical of 802
759 topology of the World Wide Web was studied, formerly random networks, power laws describe systems in which few 803
760 thought to show a bell-shaped degree distribution, as in the case hubs dominate:105 networks that are characterized by a power- 804
761 of random graphs. law degree distribution are highly nonuniform, most of the 805
762 Instead, by counting how many Web pages have exactly k nodes having only a few links. Only few nodes with a very large 806
763 links the authors showed that the distribution followed a so- number of links, which are often called hubs, hold these nodes 807
764 called power law, namely, the probability that any node is together. 808
765 connected to k other nodes is A key feature of many complex systems is their robustness, 809
pk = αk −γ which refers to the system’s ability to respond to changes in the 810
external conditions or internal organization while maintaining 811
766 where γ is the degree exponent and α is the proportionality relatively normal behavior.107 In a random network, disabling a 812
767 constant. The value of γ determines many properties of the substantial number of nodes will result in an inevitable 813
768 system. The smaller the value of γ, the more important the role functional disintegration of a network, breaking the network 814
769 of the hubs is in the network. Whereas for γ > 3 the hubs are into isolated node clusters.107 815
770 not relevant, for 2 < γ < 3 there is a hierarchy of hubs, with the Scale-free networks do not have a critical threshold for 816
771 most connected hubs being in contact with a small fraction of disintegration (percolation threshold108): they are amazingly 817
I dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX
10. Chemical Reviews Review
818 robust against accidental failures: even if 80% of randomly follows modular topological organization; this assumption has 861
819 selected nodes fail, the remaining 20% still form a compact been applied to biological neural networks, showing that the 862
820 cluster with a path connecting any two nodes.107 This is dynamic behavior of neural networks might be coordinated 863
821 because random failure is likely to affect mainly the several through different topological features,111 such as network 864
822 small degree nodes, whose removal does not disrupt the modularity and the presence of central hub nodes. A similar 865
823 networks integrity.85 This reliance on hubs, on the other hand, topology/dynamics relation seems to hold for contact net- 866
824 induces a so-called attack vulnerability: the removal of a few key works, too. As a matter of fact, allosteric “hot spots”,65 where 867
825 hubs splinters the system into small isolated node clusters.85 the motion is generalized from a local excitation to the entire 868
826 Scale-free architecture can exhibit the so-called “small word protein structure, correspond to central residue contacts, which 869
827 property”.38,104 The small word model has its roots in the were demonstrated to be crucial for efficient allosteric 870
828 observation that many real-world networks show the following communications.41,59,66,72,86 871
829 two properties: (i) the small-world effect (i.e., small average The field of relations between molecular dynamics 872
830 shortest path length) and (ii) high clustering or transitivity, trajectories and topological contact network description is a 873
831 meaning that there is a heightened probability that two vertices very important avenue of research in protein science.62−64,66,70 874
832 will be connected directly to one another if they have another
833 neighboring vertex in common. 3. APPLICATIONS
834 The former property is quantified by the characteristic path 3.1. Networks and Interactions
835 length (or average shortest path) l of the graph, while the
836 second property is computed as the clustering coefficient C. It is well-known that proteins interact among themselves and 875
837 Thus, small-world effect means that the average shortest path in with other molecules to perform their biological functions;69 876
838 the network scales logarithmically with graph size73,109,110 crucial factors in all interactions are the shape and chemical 877
properties of the pockets located on protein surfaces, which 878
l ∝ log(N ) show high affinity to binding sites. In a recent work,112 the 879
analysis of topological properties of the pocket similarity 880
839 where N is the number of nodes. network demonstrated that highly connected pockets (hubs) 881
840 PCNs were analyzed as for their scale-free properties, in generate similar concavity patterns on different protein surfaces. 882
841 order to identify crucial binding sites.43,59 The small-world These similarities go hand-in-hand with similar biological 883
842 behavior of protein structure networks was shown for the first functions that imply similar pockets.112 In addition, they found 884
843 time by Vendruscolo et al.43 and later confirmed in several that maximum connected components in the pocket similarity 885
844 works.38,75 As we stretched before, it was shown that small- networks have a small-world and scale-free scaling. The analysis 886
845 world behavior of an inter-residue contact graph is conditioned of the physicochemical features of hub pockets leads to the 887
846 by the backbone connectivity.104 investigation of more functional implications from the similarity 888
847 According to both,59,104 PCNs are not “pure small-world” network model, which provided new insights into structural 889
848 networks, given that no explicit hub is present, so they must be genomics and have great potential for applications in functional 890
849 considered as “a class of network in its own”, generated by the genomics.113 The future purpose is to develop a classification 891
850 very peculiar constraint to mantain a continuous (covalent) method to divide similar pockets into small groups and 892
851 backbone joining the nodes in a fixed sequence.59,104 afterward to compile this evolutionary information into a library 893
852 Nevertheless, the most important feature of small-world of functional templates. 894
853 architecture, i.e., the presence of shortcuts allowing for an This work delineates a possible link between network wiring 895
854 efficient signal transmission at long distance, is present in PCNs and common function of utmost interest for the development 896
855 and it is the very basis of their physiological role (allostery, of contact-based meaningful formulas. By briefly describing 897
f9 856 dynamical properties, folding rate, etc.) (Figure 9). direct translation of graph theoretical descriptors into mean- 898
857 In this respect, it is relevant to go more in-depth into the link ingful protein functional properties, we gave a proof-of-concept 899
858 existing between a given topology and the dynamical behavior it of the general relevance of the proposed formalism. Now we 900
859 can host. As a matter of fact, according to a pattern-based will go more in-depth into some of these topology−function 901
860 computational approach,111 modular dynamic organization relations, but a leading leitmotiv can be already stated: the 902
structure−function link passes through a topological bottle- 903
neck, the contact network, that allows for a consistent and very 904
efficient formalism to be applied to the study of macro- 905
molecules. 906
3.2. Protein Structure Classification
Proteins can be considered as modular geometric objects 907
composed of blocks, so allowing for a peptide-fragment-based 908
partition.114 For instance, it is well-known that globular 909
proteins are made up of regular secondary structures (α-helices 910
and β-strands) and nonregular secondary regions, called loops, 911
that join regular secondary structures and lack the regularity of 912
torsion angles for consecutive residues; actually, many families 913
of proteins evolved to perform multiple functions, with 914
variations in loop regions on a relatively conserved secondary 915
Figure 9. An example of a small-world network: most nodes are linked structure framework. Considering this, Tendulkar et al.114 916
only to their immediate neighbors, while few edges generate shortcuts developed an unconventional scheme of loops and secondary 917
between distant regions of the network. structure classification: the clustering of the peptide fragments 918
J dx.doi.org/10.1021/cr3002356 | Chem. Rev. XXXX, XXX, XXX−XXX