SlideShare a Scribd company logo
1 of 16
SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network
for Efficient and Generalizable
Protein-Protein Interaction Prediction
Ziyuan Zhao1,2,*, Peisheng Qian1,*, Xulei Yang1,+ , Zeng Zeng3 , Cuntai Guan2 , Wai Leong Tam4 , Xiaoli Li1,2
Presenter: Ziyuan Zhao, Peisheng Qian
1 Institute for Infocomm Research (I2R), A*STAR, Singapore
2 School of Computer Science and Engineering (SCSE), Nanyang Technological University, Singapore
3School of Microelectronics, Shanghai University, China
4 Genome Institute of Singapore (GIS), A*STAR, Singapore
Paper ID: 2877
2
Challenges in Protein-Protein Interaction Prediction
- Protein-protein Interactions (PPIs) are central to various cellular functions
and processes.
- Label Scarcity: PPIs need to be annotated and may not be available.
- Domain Shift: Models trained on one domain can suffer tremendous
performance degradation when evaluated on another domain.
3
Improving Efficiency and generalization in PPI prediction
- Machine learning (ML) based, deep learning (DL) based and Graph
Neural Network (GNN) based methods have been investigated.
- However, dealing with imperfect data for improving model efficiency and
generalization in PPI prediction remains underexplored.
4
SemiGNN-PPI: Self-ensembling Multi-graph Neural Network
- Multi-graph encoding
- GNN + Mean Teacher
- Graph consistency constraints
5
Multi-Graph Encoding
- PPI Graph: proteins and PPIs.
- Label graph: PPI types and their correlations.
- Protein-Graph Encoding (PGE) aggregates
representations from neighboring proteins.
- Label-Graph Encoding (LGE) learns inter-dependent
classifiers.
- Multi-Graph Based Classifier applies the
learned classifiers from LGE to
representations from PGE for the PPI
prediction scores.
6
Self-ensemble Graph Learning
- We adopt mean teaching with graph data augmentation.
- Edge Manipulation (EM): for connectivity variations,
we randomly replace a certain percentage of edges.
- Node Manipulation (NM): for attribute missing, we
randomly remove node features, with zero masking.
- We construct two augmented graph views for the student
and teacher networks for consistent predictions.
7
Graph Consistency Constraints
- We model the fine-grained structural
protein-protein relations in the feature
embedding space [Ma et al., 2022].
- Edge matching:
- Student embedding graph
- Teacher embedding graph
- Consistent instance-wise correlations
- Edge matching loss
Yuchen Ma, Yanbei Chen, and Zeynep Akata. Distilling knowledge from self-supervised teacher by embedding graph alignment.
In 33rd British Machine Vision Conference. BMVA Press, 2022
Student Protein
Encoding
Teacher Protein
Encoding
8
Graph Consistency Constraints
- Node matching:
- Edge embedding graph
- Aligning encoding of the same protein
- node matching loss
- Overall loss function
Yuchen Ma, Yanbei Chen, and Zeynep Akata. Distilling knowledge from self-supervised teacher by embedding graph alignment.
In 33rd British Machine Vision Conference. BMVA Press, 2022
Student Protein
Encoding
Teacher Protein
Encoding
9
Datasets and Settings
- 3 datasets, STRING, SHS148k, and SHS27k.
- 7 PPI types: activation, binding, catalysis, expression, inhibition, post-
translational modification (ptmod), and reaction.
- Random, breath-first search (BFS), and depth-first search (DFS)
partitions [Lv et al., 2021].
- Evaluation metric: F1.
Guofeng Lv, Zhiqiang Hu, Yanguang Bi, and Shaoting Zhang. Learning unknown from correlations: Graph neural network for inter-novel-protein interaction prediction, Proceedings of the Thirtieth
International Joint Conference on Artificial Intelligence, IJCAI-21, pages 3677–3683
10
Comparison with Baselines
- Comparing with Machine Learning (ML), Deep Learning (DL) approaches
and GNN-PPI (a strong graph-based baseline).
- We outperform baselines by a clear margin in all datasets and partition
schemes.
11
Experiments under Label Scarcity
- We use 5%, 10% and 20% labels in the train set.
- Our method achieves better performance under all scenarios with
different datasets, label ratios, and partition schemes.
12
Experiments under Domain Shift
- The model is trained and tested in 3
settings.
- Domain Generalization
(DG)
- Inductive Domain Adaptation
(IDA)
- Transductive domain adaptation
(TDA)
13
Inter-novel-protein Interaction Prediction
- In the labeled train set
- BS subset (both proteins of the PPI are present).
- ES subset (either one protein of the PPI is present).
- NS subset (neither of the proteins is present).
14
Conclusions
- We identified 2 challenges in PPI prediction, label scarcity and domain
shift. We addressed them with a novel SemiGNN-PPI for efficient and
generalizable multi-type PPI prediction.
- To enhance generalization capability, we constructed and processed
graphs at protein and label levels.
- To leverage unlabeled PPI data, We integrated GNN into Mean Teacher
and designed multiple graph consistency constraints.
- Experiment results validated the effectiveness of SemiGNN-PPI.
15
Acknowledgement
This research was funded by Competitive Research Programme “NRF-
CRP22-2019-0003”, National Research Foundation Singapore, and partially
supported by A*STAR core funding.
Ziyuan Zhao zhao_ziyuan@i2r.a-star.edu.sg
Peisheng Qian qian_peisheng@i2r.a-star.edu.sg
Thank you!

More Related Content

Similar to [IJCAI 2023] SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction

AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...
AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...
AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...IAEME Publication
 
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...ssuser4b1f48
 
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaComparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaIRJET Journal
 
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...thanhdowork
 
NS-CUK Seminar: V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...
NS-CUK Seminar:  V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...NS-CUK Seminar:  V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...
NS-CUK Seminar: V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...ssuser4b1f48
 
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...ssuser4b1f48
 
Deep Graph Contrastive Representation Learning.pptx
Deep Graph Contrastive Representation Learning.pptxDeep Graph Contrastive Representation Learning.pptx
Deep Graph Contrastive Representation Learning.pptxssuser2624f71
 
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHM
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHMBACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHM
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHMcscpconf
 
Comparison of Neural Network Training Functions for Hematoma Classification i...
Comparison of Neural Network Training Functions for Hematoma Classification i...Comparison of Neural Network Training Functions for Hematoma Classification i...
Comparison of Neural Network Training Functions for Hematoma Classification i...IOSR Journals
 
Adaptive modified backpropagation algorithm based on differential errors
Adaptive modified backpropagation algorithm based on differential errorsAdaptive modified backpropagation algorithm based on differential errors
Adaptive modified backpropagation algorithm based on differential errorsIJCSEA Journal
 
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...Joonhun Lee
 
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...IRJET Journal
 
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...IRJET Journal
 
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...sipij
 
2018_Enhanced SRGAN.pdf
2018_Enhanced SRGAN.pdf2018_Enhanced SRGAN.pdf
2018_Enhanced SRGAN.pdfSekharSankuri1
 
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...thanhdowork
 
Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013Pedro Lopes
 
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...Ziyuan Zhao
 
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...ssuser4b1f48
 

Similar to [IJCAI 2023] SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction (20)

AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...
AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...
AN IMPROVED METHOD FOR IDENTIFYING WELL-TEST INTERPRETATION MODEL BASED ON AG...
 
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...
NS-CUK Seminar: S.T.Nguyen, Review on "Improving Graph Neural Network Express...
 
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of GlaucomaComparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
Comparative Study of Pre-Trained Neural Network Models in Detection of Glaucoma
 
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...
240304_Thuy_Labseminar[SimGRACE: A Simple Framework for Graph Contrastive Lea...
 
1207.2600
1207.26001207.2600
1207.2600
 
NS-CUK Seminar: V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...
NS-CUK Seminar:  V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...NS-CUK Seminar:  V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...
NS-CUK Seminar: V.T.Hoang, Review on "Namkyeong Lee, et al. Relational Self-...
 
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...
NS-CUK Journal club: HELee, Review on "Graph embedding on biomedical networks...
 
Deep Graph Contrastive Representation Learning.pptx
Deep Graph Contrastive Representation Learning.pptxDeep Graph Contrastive Representation Learning.pptx
Deep Graph Contrastive Representation Learning.pptx
 
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHM
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHMBACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHM
BACKPROPAGATION LEARNING ALGORITHM BASED ON LEVENBERG MARQUARDT ALGORITHM
 
Comparison of Neural Network Training Functions for Hematoma Classification i...
Comparison of Neural Network Training Functions for Hematoma Classification i...Comparison of Neural Network Training Functions for Hematoma Classification i...
Comparison of Neural Network Training Functions for Hematoma Classification i...
 
Adaptive modified backpropagation algorithm based on differential errors
Adaptive modified backpropagation algorithm based on differential errorsAdaptive modified backpropagation algorithm based on differential errors
Adaptive modified backpropagation algorithm based on differential errors
 
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...
MINR: Implicit Neural Representations with Masked Image Modelling (ICCV '23 O...
 
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...
IRJET - Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence M...
 
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...
IRJET- Plant Leaf Disease Diagnosis from Color Imagery using Co-Occurrence Ma...
 
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...
ADVANCED SINGLE IMAGE RESOLUTION UPSURGING USING A GENERATIVE ADVERSARIAL NET...
 
2018_Enhanced SRGAN.pdf
2018_Enhanced SRGAN.pdf2018_Enhanced SRGAN.pdf
2018_Enhanced SRGAN.pdf
 
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...
240318_Thuy_Labseminar[Fragment-based Pretraining and Finetuning on Molecular...
 
Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013Poster_Reseau_Neurones_Journees_2013
Poster_Reseau_Neurones_Journees_2013
 
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...
[ICIP 2022] ACT-NET: Asymmetric Co-Teacher Network for Semi-Supervised Memory...
 
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...
NS - CUK Seminar: V.T.Hoang, Review on "Long Range Graph Benchmark.", NeurIPS...
 

More from Ziyuan Zhao

[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning for Medical...
[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning  for Medical...[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning  for Medical...
[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning for Medical...Ziyuan Zhao
 
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...Ziyuan Zhao
 
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...Ziyuan Zhao
 
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...Ziyuan Zhao
 
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...Ziyuan Zhao
 
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...Ziyuan Zhao
 
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...Ziyuan Zhao
 
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age EstimationZiyuan Zhao
 
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...Ziyuan Zhao
 
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...Ziyuan Zhao
 
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...Ziyuan Zhao
 
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...Ziyuan Zhao
 
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...Ziyuan Zhao
 

More from Ziyuan Zhao (13)

[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning for Medical...
[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning  for Medical...[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning  for Medical...
[IAIM 2023 - Poster] Label-efficient Generalizable Deep Learning for Medical...
 
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...
[BMVC 2022 - Spotlight] DA-CIL: Towards Domain Adaptive Class-Incremental 3D ...
 
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...
[BMVC 2022] DA-CIL: Towards Domain Adaptive Class-Incremental 3D Object Detec...
 
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...
[ICIP 2022] MMGL: Multi-Scale Multi-View Global-Local Contrastive learning fo...
 
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...
[ICIP 2022 - Poster] MMGL: Multi-Scale Multi-View Global-Local Contrastive le...
 
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...
[MICCAI 2022] Meta-hallucinator: Towards Few-Shot Cross-Modality Cardiac Imag...
 
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...
[EMBC 2022] Self-supervised Assisted Active Learning for Skin Lesion Segmenta...
 
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation
[ICME 2022] Adaptive Mean-Residue Loss for Robust Facial Age Estimation
 
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...
[MICCAI 2021] MT-UDA: Towards unsupervised cross-modality medical image segme...
 
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...
[EMBC 2021] Hierarchical Consistency Regularized Mean Teacher for Semi-superv...
 
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...
[EMBC 2021] Multi Slice Dense Sparse Learning for Efficient Liver and Tumor S...
 
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...
[ICIP 2020] SEA-Net: Squeeze-and-Excitation Attention Net for Diabetic Retino...
 
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...
[MICCAI 2021 - Poster] MT-UDA: Towards unsupervised cross-modality medical im...
 

Recently uploaded

Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘RTylerCroy
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 

Recently uploaded (20)

Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
🐬 The future of MySQL is Postgres 🐘
🐬  The future of MySQL is Postgres   🐘🐬  The future of MySQL is Postgres   🐘
🐬 The future of MySQL is Postgres 🐘
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 

[IJCAI 2023] SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction

  • 1. SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction Ziyuan Zhao1,2,*, Peisheng Qian1,*, Xulei Yang1,+ , Zeng Zeng3 , Cuntai Guan2 , Wai Leong Tam4 , Xiaoli Li1,2 Presenter: Ziyuan Zhao, Peisheng Qian 1 Institute for Infocomm Research (I2R), A*STAR, Singapore 2 School of Computer Science and Engineering (SCSE), Nanyang Technological University, Singapore 3School of Microelectronics, Shanghai University, China 4 Genome Institute of Singapore (GIS), A*STAR, Singapore Paper ID: 2877
  • 2. 2 Challenges in Protein-Protein Interaction Prediction - Protein-protein Interactions (PPIs) are central to various cellular functions and processes. - Label Scarcity: PPIs need to be annotated and may not be available. - Domain Shift: Models trained on one domain can suffer tremendous performance degradation when evaluated on another domain.
  • 3. 3 Improving Efficiency and generalization in PPI prediction - Machine learning (ML) based, deep learning (DL) based and Graph Neural Network (GNN) based methods have been investigated. - However, dealing with imperfect data for improving model efficiency and generalization in PPI prediction remains underexplored.
  • 4. 4 SemiGNN-PPI: Self-ensembling Multi-graph Neural Network - Multi-graph encoding - GNN + Mean Teacher - Graph consistency constraints
  • 5. 5 Multi-Graph Encoding - PPI Graph: proteins and PPIs. - Label graph: PPI types and their correlations. - Protein-Graph Encoding (PGE) aggregates representations from neighboring proteins. - Label-Graph Encoding (LGE) learns inter-dependent classifiers. - Multi-Graph Based Classifier applies the learned classifiers from LGE to representations from PGE for the PPI prediction scores.
  • 6. 6 Self-ensemble Graph Learning - We adopt mean teaching with graph data augmentation. - Edge Manipulation (EM): for connectivity variations, we randomly replace a certain percentage of edges. - Node Manipulation (NM): for attribute missing, we randomly remove node features, with zero masking. - We construct two augmented graph views for the student and teacher networks for consistent predictions.
  • 7. 7 Graph Consistency Constraints - We model the fine-grained structural protein-protein relations in the feature embedding space [Ma et al., 2022]. - Edge matching: - Student embedding graph - Teacher embedding graph - Consistent instance-wise correlations - Edge matching loss Yuchen Ma, Yanbei Chen, and Zeynep Akata. Distilling knowledge from self-supervised teacher by embedding graph alignment. In 33rd British Machine Vision Conference. BMVA Press, 2022 Student Protein Encoding Teacher Protein Encoding
  • 8. 8 Graph Consistency Constraints - Node matching: - Edge embedding graph - Aligning encoding of the same protein - node matching loss - Overall loss function Yuchen Ma, Yanbei Chen, and Zeynep Akata. Distilling knowledge from self-supervised teacher by embedding graph alignment. In 33rd British Machine Vision Conference. BMVA Press, 2022 Student Protein Encoding Teacher Protein Encoding
  • 9. 9 Datasets and Settings - 3 datasets, STRING, SHS148k, and SHS27k. - 7 PPI types: activation, binding, catalysis, expression, inhibition, post- translational modification (ptmod), and reaction. - Random, breath-first search (BFS), and depth-first search (DFS) partitions [Lv et al., 2021]. - Evaluation metric: F1. Guofeng Lv, Zhiqiang Hu, Yanguang Bi, and Shaoting Zhang. Learning unknown from correlations: Graph neural network for inter-novel-protein interaction prediction, Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21, pages 3677–3683
  • 10. 10 Comparison with Baselines - Comparing with Machine Learning (ML), Deep Learning (DL) approaches and GNN-PPI (a strong graph-based baseline). - We outperform baselines by a clear margin in all datasets and partition schemes.
  • 11. 11 Experiments under Label Scarcity - We use 5%, 10% and 20% labels in the train set. - Our method achieves better performance under all scenarios with different datasets, label ratios, and partition schemes.
  • 12. 12 Experiments under Domain Shift - The model is trained and tested in 3 settings. - Domain Generalization (DG) - Inductive Domain Adaptation (IDA) - Transductive domain adaptation (TDA)
  • 13. 13 Inter-novel-protein Interaction Prediction - In the labeled train set - BS subset (both proteins of the PPI are present). - ES subset (either one protein of the PPI is present). - NS subset (neither of the proteins is present).
  • 14. 14 Conclusions - We identified 2 challenges in PPI prediction, label scarcity and domain shift. We addressed them with a novel SemiGNN-PPI for efficient and generalizable multi-type PPI prediction. - To enhance generalization capability, we constructed and processed graphs at protein and label levels. - To leverage unlabeled PPI data, We integrated GNN into Mean Teacher and designed multiple graph consistency constraints. - Experiment results validated the effectiveness of SemiGNN-PPI.
  • 15. 15 Acknowledgement This research was funded by Competitive Research Programme “NRF- CRP22-2019-0003”, National Research Foundation Singapore, and partially supported by A*STAR core funding.
  • 16. Ziyuan Zhao zhao_ziyuan@i2r.a-star.edu.sg Peisheng Qian qian_peisheng@i2r.a-star.edu.sg Thank you!

Editor's Notes

  1. Good afternoon session chairs and everyone here today. My name is Peisheng. Today I will present our paper SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction.
  2. Protein-protein Interactions are central to many cellular functions and processes. The PPI prediction is a classification problem where we try to predict the classes of the interactions between to 2 proteins. This is important because PPIs have significant implications for drug development and disease diagnosis. However, in real-world scenarios, PPI prediction is affected by various factors, such as label scarcity and domain shift. In the label scarcity scenario, PPI need to be annotated from experiments and only a small portion of them can be used for training. The lack of labels can be a significant bottleneck for PPI prediction. In the domain shift scenario, most existing methods are only developed and validated using in-distribution data and they receive severe performance degradation for unseen data with different distributions. Therefore, label scarcity and domain shift are 2 challenges in PPI prediction.
  3. To deal with label scarcity, we aim at improving the data efficiency. To alleviate domain shift, we aim at enhancing the generalization capability for PPI prediction. Computational approaches for PPI prediction include machine learning based and deep learning based methods. Since PPI can naturally be formulated as graphs with proteins as nodes and interactions as edges, graph neural networks have also been investigated. However, dealing with imperfect data for improving model efficiency and generalization remains a vital but underexplored issue.
  4. In our approach, For generalizable PPI prediction, we use multi-graph encoding to model protein correlations and label dependencies, in which, we construct graphs to learn correlations between proteins and label dependencies simultaneously. For data efficient PPI prediction, we advance GNN with Mean Teacher to explore unlabeled data for self-ensemble graph learning. Moreover, we apply multiple graph consistency constraints for regularization in self ensemble learning. We will go through the 3 points in the following slides.
  5. The proposed multi-graph encoding is based on 2 graphs. For the PPI graph, the protein are nodes and PPIs are edges. For the label graph, the PPI types are nodes and correlations between the PPI types are edges. To obtain protein graph encodings, we use graph neural networks to aggregate representations from neighboring proteins in the PPI graph. To learn correlations among different types of interactions, we learn inter-dependent classifiers using Graph Convolutional Network. Then, we apply the learned classifiers from label graph encoding to the learned representations from protein graph encoding and obtain the predicted classification scores.
  6. Next, to leverage unlabeled data, we adopt the mean teaching architecture. To facilitate self-ensemble graph learning, we use two data augmentation methods at both the edge and node level. The edge manipulation aims to improve the robustness against connectivity variations. We randomly replace a certain percentage of edges in the input to the models since some PPIs could be unidentified or wrongly identified. The node manipulation aims to improve the robustness against missing attributes, we randomly mask node features with zeros and feed them into the models. And we expect the model to effectively learn features even in the presence of missing attribute information. We use edge and node manipulations to construct two augmented graph views to feed the student and teacher networks separately, and encourage them to generate consistent predictions.
  7. Aside from consistency in the prediction space. We also want to model the fine-grained structural protein-protein relations in the feature embedding space. To achieve this, we use edge matching and node matching. For edge matching, We calculate all pairwise Pearson’s correlation coefficient between nodes in the same batch from the student network and call it the student embedding graph. Similarly, we construct the teacher embedding graph. Then we enforce consistent instance-wise correlations using the edge matching loss. In the loss function, Gse refers to the student embedding graph, Gte refers to the teacher embedding graph, and Adj refers to the adjacency matrix.
  8. We also use node matching as another constraint. For node matching, We formulate the edge embedding graph by calculating all pairwise Pearson’s correlation coefficient between student encoding and teacher encoding in the same batch. We align the encoding of the same protein from the 2 networks in a node matching loss. In the loss function, Gste is the edge embedding graph, I is the identity matrix and diag is the operation to keep only diagonal values in the matrix and set the rest to 0. The overall loss is a weighted sum of the supervised loss, consistency loss, node and edge matching loss.
  9. Experiments were conducted on 3 datasets, STRING, SHS148k, and SHS27k. The PPIs are annotated with 7 types. Each PPI is labeled with at least one of them. We follow existing partition algorithms, and use random, breath-first search, and depth-first search on protein nodes to create test sets with 20% of data. The rest of the data are used as the train set. The BFS and DFS create test data with more unseen proteins, which are more challenging scenarios. We use F1 score as the evaluation metric.
  10. We compare with Machine Learning and Deep Learning baselines. Particularly, GNN-PPI is a strong graph-based baseline for multi-class PPI prediction. In this table, our method outperforms baselines by a clear margin in all datasets and all partition schemes.
  11. Next, We use 5%, 10% and 20% labels in the train set to simulate the label scarcity scenario. In this case, GNN-PPI receives severe performance degradation with fewer labels. In comparison, our method achieves better performance under all scenarios with different datasets, label ratios, and partition schemes.
  12. To access the generalization capability of the proposed method, we test the model in 3 evaluation settings: For Domain Generalization: The model does not have access to the trainset-heterologous dataset, and tested on the unseen dataset. For Inductive Domain Adaptation: The model has access to unlabeled training data in the trainset-heterologous dataset. For Transductive Domain Adaptation: The model has access to the whole un-labeled trainset-heterologous dataset. Our method outperforms GNN-PPI in all of the 3 settings.
  13. Next, we analyze model performance on inter-novel-protein interaction, where the proteins could be present or absent in the labeled trainset. The ES and NS subsets are more challenging because the PPI to predict is between proteins that the model did not see during training. In this table, our method outperforms GNN-PPI in most subsets.
  14. In conclusion, We identified 2 challenges in PPI prediction, label scarcity and domain shift. We addressed them with a novel self-ensembling multi-graph neural network for efficient and generalizable multi-type PPI prediction. To enhance generalization capability, we constructed and processed graphs at protein and label levels. To leverage unlabeled PPI data, we integrated GNN into Mean Teacher and used multiple graph consistency constraints to align feature embeddings. Finally, experimental results proved the effectiveness of our approach.
  15. The research was funded by National Research Foundation Singapore, and was partially supported by A*STAR core funding. We would also like to say thank you for all collaborators from NTU, Shanghai University and Genome Institute of Singapore.
  16. That’s all and thank you for listening!