2016 09-16-fairdom
•
1 like•2,628 views
Keynote talk given at Fairdom User meeting http://fair-dom.org/communities/users/barcelona-2016-first-user-meeting/ . I begin by summarising how we apply molecular approaches to understand social behaviour in ants. Subsequently, I give an overview of the data-handling challenges the genomic bioinformatics community faces. Finally, I give an overview of some of the tools and approaches my lab have developed to help us get things done better, faster, more reliably and more reproducibly.
Recommended
Recommended
More Related Content
Viewers also liked
Viewers also liked (10)
Similar to 2016 09-16-fairdom
More from Yannick Wurm
More from Yannick Wurm (20)
Recently uploaded
Recently uploaded (20)
2016 09-16-fairdom
- 1. A major transition in social evolution & some data tidbits y.wurm@qmul.ac.uk https://wurmlab.github.io
- 2. © Alex Wild & others
- 4. © National Geographic Atta leaf-cutter ants
- 5. © National Geographic Atta leaf-cutter ants
- 6. © National Geographic Atta leaf-cutter ants
- 8. Oecophylla Weaver ants © ameisenforum.de
- 10. © ameisenforum.de Oecophylla Weaver ants
- 12. Tofilski et al 2008 Forelius pusillus
- 13. Tofilski et al 2008 Forelius pusillus hides the nest entrance at night
- 14. Tofilski et al 2008 Forelius pusillus hides the nest entrance at night
- 15. Tofilski et al 2008 Forelius pusillus hides the nest entrance at night
- 16. Tofilski et al 2008 Forelius pusillus hides the nest entrance at night
- 17. Avant Workers staying outside die « preventive self-sacrifice » Tofilski et al 2008 Forelius pusillus hides the nest entrance at night
- 18. Dorylus driver ants: ants with no home © BBC
- 20. Animal biomass (Brazilian rainforest) from Fittkau & Klinge 1973 Other insects Amphibians Reptiles Birds Mammals Earthworms Spiders Soil fauna excluding earthworms, ants & termites Ants & termites
- 21. Well-studied: • behavior • morphology • evolutionary context
- 22. Well-studied: • behavior • morphology • evolutionary context • ecology Genetic basis?
- 25. REPORTS onMarch12,2013www.sciencemag.orgDownloadedfrom Solenopsis invicta fire ants are a big problem! verywellstudied! Ascunceetal2011
- 26. Solenopsis invicta fire ant: two social forms •1 large queen •Independent founding •Highly territorial •Many sizes of workers •2-100 smaller queens •Dependent founding •No inter-colony aggression •All workers similar size Single-queen form: Multiple-queen form:
- 27. Fire ants + Population genetics:Allozyme screen Ken Ross “starch gel”+ 1 2 3 L. Keller
- 28. Allozyme screen Social form associated to Gp-9 locus Frequency of the most common allele Locus! 0.3! 0.4! 0.5! 0.6! 0.7! 0.8! 0.9! 1.0! Single queen!Multiple queen! Est-6!Est-4! G 3pdh-1!C a-4!Pgm -4!Ddh-1!Pro-5! Pgm -3!Acoh-5!acoh-1!A cy-1!Pgm -1!Aat-2!Gp-9! Ken Ross and colleagues Laurent Keller and colleagues
- 29. Single queen form Multiple queen form Ken Ross and colleagues Laurent Keller and colleagues Social form completely associated to Gp-9 locus
- 30. bbbbBB BB Bb bb Ken Ross and colleagues Laurent Keller and colleagues Single queen form Multiple queen form Social form completely associated to Gp-9 locus (>15% )(< 5% )
- 31. bbBB BB Bb x Gp-9 bb females rare Ken Ross and colleagues Laurent Keller and colleagues Single queen form Multiple queen form Social form completely associated to Gp-9 locus (>15% )(< 5% )
- 32. BB BB Bb Ken Ross and colleagues Laurent Keller and colleagues Single queen form Multiple queen form Social form completely associated to Gp-9 locus (>15% )(< 5% )
- 33. BB BB Bb x Ken Ross and colleagues Laurent Keller and colleagues Single queen form Multiple queen form Social form completely associated to Gp-9 locus (>15% )(< 5% )
- 34. BB BB Bb x x Ken Ross and colleagues Laurent Keller and colleagues Social form completely associated to Gp-9 locus Single queen form Multiple queen form (>15% )(< 5% )
- 35. BB BB Bb x x x Ken Ross and colleagues Laurent Keller and colleagues Single queen form Multiple queen form (>15% )(< 5% ) Social form completely associated to Gp-9 locus
- 36. • Is this gene the single überregulator? Social form completely associated to Gp-9 locus
- 37. • Is this gene the single überregulator? maybe 1/14th of the genome? •Only 14 allozyme markers Locus! 0.3! 0.4! 0.5! 0.6! 0.7! 0.8! 0.9! 1.0! Single queen!Multiple queen! Est-6!Est-4! G 3pdh-1!C a-4!Pgm -4!Ddh-1!Pro-5! Pgm -3!Acoh-5!acoh-1!A cy-1!Pgm -1!Aat-2!Gp-9! Social form completely associated to Gp-9 locus
- 38. This changes everything. Any lab can sequence anything! http://genome.gov/sequencingcosts
- 40. Bb unfertilised eggs haploid ♂ Gp-9 B Gp-9 b Gp-9 B Gp-9 b Gp-9 b Gp-9 B 38 B♂ & 38 b♂ RAD genotyping
- 41. Identify polymorphism individual x locus genotype table RAD genotyping: sequencing the same 0.01% of the genome in many individuals A B C D E F L1 A C A A C C L2 G G T - T G L3 - A G A - G L4 C - - G G C L5 T T C T C - L6 G A A - - G 2419loci 38 B & 38 b 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20+ Amount of variance explained per principal component %VarianceExplained 051015202530 12.7% 6.1% 5.4% 4.8% 4.7% 3.9% 3.5% 3.2% 3.1% 2.9% 2.8% 2.6% 2.4% 2.3% 2.2% 2.0% 1.9% 1.7% 1.6% 30.2% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20+ Amount of variance explained per principal component %VarianceExplained 051015202530 PCA: Principal Component Analysis
- 42. pc: 2 % variance: 6.073 pc:3%variance:5.441 -0.2 -0.1 0.0 0.1 0.2 -0.2 -0.1 0.0 0.1 0.2 pc: 2 % variance: 6.073 pc:3%variance:5.441 -0.2 -0.1 0.0 0.1 0.2 -0.2 -0.1 0.0 0.1 0.2 Principal Components: PC2 vs PC3 Gp-9 B Gp-9 b
- 43. pc: 1 % variance: 12.666 pc:2%variance:6.073 -0.2 -0.1 0.0 0.1 0.2 -0.10 -0.05 0.00 0.05 0.10 0.15 Principal Components: PC1 vs PC2 pc: 1 % variance: 12.666 pc:2%variance:6.073 -0.2 -0.1 0.0 0.1 0.2 -0.10 -0.05 0.00 0.05 0.10 0.15 Gp-9 B ♂ Gp-9 b ♂
- 45. • Is this gene the single überregulator? maybe 1/14th of the genome? •Only 14 allozyme markers Social form completely associated to Gp-9 locus BB BB Bb Single queen form Multiple queen form (>15% )(< 5% ) x xx ✖ ✔ Locus! 0.3! 0.4! 0.5! 0.6! 0.7! 0.8! 0.9! 1.0! Single queen!Multiple queen! Est-6!Est-4! G 3pdh-1!C a-4!Pgm -4!Ddh-1!Pro-5! Pgm -3!Acoh-5!acoh-1!A cy-1!Pgm -1!Aat-2!Gp-9!
- 46. Sex chromosomes X Y Gp-9 B Gp-9 b SB Sb ? 1.Why non-recombining? “Social chromosomes” = supergene 2.Are SB and Sb differentiated? 3.What are the differences?
- 48. SBSB SBSb Single queen form Multiple queen form SBSB SB Sb Single queen colony Multiple queen colony SBSB SB Sb Single queen colony Multiple queen colony Summary: Fire ants have two colony types Summary: this is determined by a pair of social chromosomes
- 50. Research themes • Biomedical approaches • International population genomics surveys • Monitoring via sequencing • Major social transitions » social chromosomes » convergence » eusociality, queen number, parasitism... • 100-fold intra-specific variation in lifespan • Strengths of selection • Candidate genes/pathway Pollinator health Genome evolution Social evolution Modern bioinformatics tools & approaches (some at https://wurmlab.github.io )
- 52. SequenceServer
- 53. “Can you BLAST this for me?” BLAST But: •convoluted interface •challenging on custom data Antgenomes.org SequenceServer BLAST made easy is the most commonly used tool: >100,000 citations
- 54. http://www.sequenceserver.com/ If no config file:Asks interactive setup questions. If needed: Downloads BLAST binaries If needed: Formats FASTA into BLAST database. 1. Installing gem install sequenceserver ### Launched SequenceServer at: http://0.0.0.0:4567 2. Launch sequenceserver Demo Anurag Priyam - @yeban
- 55. http://www.sequenceserver.com/ Anurag Priyam @yeban
- 56. http://www.sequenceserver.com/ Anurag Priyam @yeban
- 57. Bionode
- 58. Timewasters • Client vs server-side code. • Workflows stalling (data download, cluster queues…) • Fragmented efforts - having to learn additional languages for specific tools + project- specific needs Bionode BrunoVieira @bmpvieira
- 59. Philosophy for flexibility Modules should: •(also) work in the web browser (when possible) •(also) work in the command-line •support streaming input/output gittergitter join chatjoin chat http://bionode.io BrunoVieira @bmpvieira
- 60. Difficulty writing scalable, reproducible and complex bioinformatic pipelines. Solution: Node.js everywhereStreams var ncbi = require('bionode-ncbi') var tool = require('tool-stream') var through = require('through2') var fork1 = through.obj() var fork2 = through.obj() ncbi .search('sra', 'Solenopsis invicta') .pipe(fork1) .pipe(dat.reads) fork1 .pipe(tool.extractProperty('expxml.Biosample.id')) .pipe(ncbi.search('biosample')) .pipe(dat.samples) fork1 .pipe(tool.extractProperty('uid')) .pipe(ncbi.link('sra', 'pubmed')) Node/Bionode for complex pipelines @bmpvieira
- 62. #"Get"descriptions"for"papers"related"to"SRA"search! bionode!ncbi!search!sra!Solenopsis!invicta!|!! !!!!!!!!tool3stream!extractProperty!uid!|!! !!!!!!!!bionode!ncbi!link!sra!pubmed!|!! !!!!!!!!tool3stream!extractProperty!destUID!|! !!!!!!!!bionode!ncbi!search!pubmed #"Get"URL"of"Solenopsis"invicta"genome" bionode3ncbi!urls!assembly!Solenopsis!invicta!|!json|!grep!genomic.fna! !http://ftp.ncbi.nlm.nih.gov/genomes/all/GCA_000188075.1_Si_gnG/GCA_000188075.1_Si_gnG_genomic.fna.gz http://bionode.io in the terminal #"Get"all"FASTQ"of"Arthropod"short"reads" bionode3ncbi!download!sra!arthropoda!|!bionode3sra!fastq3dump!3 #"Get"all"GFF"of"bacterial"genome"annotations" bionode3ncbi!download!gff!bacteria! @bmpvieira
- 63. BrunoVieira @bmpvieira Philosophy for flexibility Modules should: •(also) work in the web browser (when possible) •(also) work in the command-line •support streaming input/output Modules: •decentralised management. •small - just do one thing well. •few strict rules, but some strong recommendations (style, interfaces etc). gittergitter join chatjoin chat
- 64. BrunoVieira @bmpvieira Contributors gittergitter join chatjoin chat YOU? BioJS for visualisation Bionode for data handling
- 66. Baby steps towards improved efficiency, robustness & reproducibility
- 68. BIG
- 71. Geoffrey Chang: Crystallographer • Beckman FoundationYoung Investigator Award • Presidential Early Career Award Journal of Molecular Biology (2003) Chang. Structure of MsbA from Vibrio cholera: a multidrug resistance ABC transporter homolog in a closed conformation. PNAS (2004) Ma & Chang. Structure of the multidrug resistance efflux transporter EmrE from Escherichia coli. Science (2005) Reyes & Chang. Structure of the ABC transporter MsbA in complex with ADP vanadate and lipopolysaccharide. Science (2005) Pornillos et al. X-ray structure of the EmrE multidrug transporter in complex with a substrate. Science (2001) Chang & Roth. Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters. Science (2001) Chang & Roth.
- 72. earch Institute in next year, in a cer- Chang received a Award rs, the young ated a apers ctures ded in into a Swiss per in bt on a group cience gated, scover ispro- mns of density m had ucture. d used energy from adenosine triphosphate to trans- port molecules across cell membranes. These so-called ABC transporters perform many determination was at the root o cess: “He has an incredible d ethic. He really pushed the fie of getting things to no one else had be Chang’s data are go but the faulty so everything off. Ironically, anoth doc in Rees’s lab, K exposed the mistake tember issue of Na now at the Swiss F ofTechnology in Zu the structure of anA calledSav1866from aureus. The structur cally—and unexpe ent from that of pulling up Sav186 MsbA from S. typh computer screen, L realized in minutes structurewasinvert the “hand” of a mol Flipping fiasco. The structures of MsbA (purple) and Sav1866 (green) overlap little (left) until MsbA is inverted (right). California.The next year, in a cer- e White House, Chang received a l Early Career Award ts and Engineers, the ghest honor for young . His lab generated a high-profile papers e molecular structures proteins embedded in nes. e dream turned into a In September, Swiss published a paper in cast serious doubt on a cture Chang’s group ed in a 2001 Science en he investigated, horrified to discover madedata-analysispro- ipped two columns of ng the electron-density which his team had final protein structure. ly, his group had used m to analyze data for port molecules across cell membranes. These so-called ABC transporters perform many cess: “He has an ethic. He really p of get no on Chan but t every Iro doc in expos temb now a ofTec the str called aureu cally— ent f pullin MsbA comp realiz struct the “h a cha Flipping fiasco. The structures of MsbA (purple) and Sav1866 (green) overlap little (left) until MsbA is inverted (right). Sav1866 Dawson & Locher (2006) Nature Science(2001)Chang&Roth.Science (2001) Chang & Roth. Comparison with 3D structure of ortholog Science (2001) Chang & Roth.
- 73. http://wurmlab.github.io LETTERS I BOOKS I POLICY FORUM I EDUCATION FORUM I PERSPECTIVES 1878 1880 1882 LETTERS edited by Etta Kavanagh Retraction WE WISH TO RETRACT OUR RESEARCH ARTICLE “STRUCTURE OF MsbA from E. coli:A homolog of the multidrug resistanceATP bind- ing cassette (ABC) transporters” and both of our Reports “Structure of the ABC transporter MsbA in complex with ADP•vanadate and lipopolysaccharide”and“X-raystructureoftheEmrEmultidrugtrans- porter in complex with a substrate” (1–3). The recently reported structure of Sav1866 (4) indicated that our MsbA structures (1, 2, 5) were incorrect in both the hand of the struc- ture and the topology. Thus, our biological interpretations based on these inverted models for MsbA are invalid. Anin-housedatareductionprogramintroducedachangeinsignfor anomalous differences.This program, which was not part of a conven- tional data processing package, converted the anomalous pairs (I+ and I-) to (F- and F+), thereby introducing a sign change. As the diffrac- tion data collected for each set of MsbA crystals and for the EmrE crystals were processed with the same program, the structures reported in (1–3, 5, 6) had the wrong hand. The error in the topology of the original MsbA structure was a con- sequence of the low resolution of the data as well as breaks in the elec- tron density for the connecting loop regions. Unfortunately, the use of the multicopy refinement procedure still allowed us to obtain reason- able refinement values for the wrong structures. The Protein Data Bank (PDB) files 1JSQ, 1PF4, and 1Z2R for MsbA and 1S7B and 2F2M for EmrE have been moved to the archive of obsolete PDB entries. The MsbA and EmrE structures will be recalculated from the original data using the proper sign for the anom- alous differences, and the new Ca coordinates and structure factors will be deposited. We very sincerely regret the confusion that these papers have caused and, in particular, subsequent research efforts that were unpro- ductive as a result of our original findings. GEOFFREY CHANG, CHRISTOPHER B. ROTH, CHRISTOPHER L. REYES, OWEN PORNILLOS, YEN-JU CHEN, ANDY P. CHEN Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA 92037, USA. References 1. G. Chang, C. B. Roth, Science 293, 1793 (2001). 2. C. L. Reyes, G. Chang, Science 308, 1028 (2005). 3. O. Pornillos, Y.-J. Chen, A. P. Chen, G. Chang, Science 310, 1950 (2005). 4. R. J. Dawson, K. P. Locher, Nature 443, 180 (2006). 5. G. Chang, J. Mol. Biol. 330, 419 (2003). 6. C. Ma, G. Chang, Proc. Natl. Acad. Sci. U.S.A. 101, 2852 (2004). MsbA from E. coli:A homolog of the multidrug resistanceATP bind- ing cassette (ABC) transporters” and both of our Reports “Structure of the ABC transporter MsbA in complex with ADP•vanadate and lipopolysaccharide”and“X-raystructureoftheEmrEmultidrugtrans- porter in complex with a substrate” (1–3). The recently reported structure of Sav1866 (4) indicated that our MsbA structures (1, 2, 5) were incorrect in both the hand of the struc- ture and the topology. Thus, our biological interpretations based on these inverted models for MsbA are invalid. Anin-housedatareductionprogramintroducedachangeinsignfor anomalous differences.This program, which was not part of a conven- tional data processing package, converted the anomalous pairs (I+ and I-) to (F- and F+), thereby introducing a sign change. As the diffrac- tion data collected for each set of MsbA crystals and for the EmrE crystals were processed with the same program, the structures reported in (1–3, 5, 6) had the wrong hand. The error in the topology of the original MsbA structure was a con- sequence of the low resolution of the data as well as breaks in the elec- 1860 Untilrecently,GeoffreyChang’scareerwason a trajectory most young scientists only dream about. In 1999, at the age of 28, the protein crystallographer landed a faculty position at the prestigious Scripps Research Institute in San Diego, California.The next year, in a cer- emony at the White House, Chang received a Presidential Early Career Award for Scientists and Engineers, the country’s highest honor for young researchers. His lab generated a stream of high-profile papers detailing the molecular structures of important proteins embedded in cell membranes. Then the dream turned into a nightmare. In September, Swiss researchers published a paper in Nature that cast serious doubt on a protein structure Chang’s group had described in a 2001 Science paper. When he investigated, Chang was horrified to discover thatahomemadedata-analysispro- 2001 Science paper, which described the struc- tureofaproteincalledMsbA,isolatedfromthe bacterium Escherichia coli. MsbA belongs to a huge and ancient family of molecules that use energy from adenosine triphosphate to trans- port molecules across cell membranes. These so-called ABC transporters perform many Sciences and EmrE, a differ Crystalliz five membra was an incred postdoc advis nia Institute o proteins are a because they ously diffic needed for x- determination cess: “He has ethic. He real of no Ch bu ev do ex tem no of the cal au ca en pu A Scientist’s Nightmare: Software Problem Leads to Five Retractions SCIENTIFIC PUBLISHING
- 74. ! Geoffrey Chang • Beckman FoundationYoung Investigator Award • Presidential Early Career Award Science (2001) Chang & Roth. Structure of MsbA from E. coli: a homolog of the multidrug resistance ATP binding cassette (ABC) transporters. Journal of Molecular Biology (2003) Chang. Structure of MsbA from Vibrio cholera: a multidrug resistance ABC transporter homolog in a closed conformation. PNAS (2004) Ma & Chang. Structure of the multidrug resistance efflux transporter EmrE from Escherichia coli. Science (2005) Reyes & Chang. Structure of the ABC transporter MsbA in complex with ADP vanadate and lipopolysaccharide. Science (2005) Pornillos et al. X-ray structure of the EmrE multidrug transporter in complex with a substrate. 1860 Untilrecently,GeoffreyChang’scareerwason a trajectory most young scientists only dream about. In 1999, at the age of 28, the protein crystallographer landed a faculty position at the prestigious Scripps Research Institute in San Diego, California.The next year, in a cer- emony at the White House, Chang received a Presidential Early Career Award for Scientists and Engineers, the country’s highest honor for young researchers. His lab generated a stream of high-profile papers detailing the molecular structures of important proteins embedded in cell membranes. Then the dream turned into a nightmare. In September, Swiss researchers published a paper in Nature that cast serious doubt on a protein structure Chang’s group had described in a 2001 Science paper. When he investigated, Chang was horrified to discover thatahomemadedata-analysispro- 2001 Science paper, which described the struc- tureofaproteincalledMsbA,isolatedfromthe bacterium Escherichia coli. MsbA belongs to a huge and ancient family of molecules that use energy from adenosine triphosphate to trans- port molecules across cell membranes. These so-called ABC transporters perform many Sciences and EmrE, a differ Crystalliz five membra was an incred postdoc advis nia Institute o proteins are a because they ously diffic needed for x- determination cess: “He has ethic. He real of no Ch bu ev do ex tem no of the cal au ca en pu A Scientist’s Nightmare: Software Problem Leads to Five Retractions SCIENTIFIC PUBLISHING
- 75. http://wurmlab.github.io This is costly For: •the individual •collaborators •the institution •1000s of researchers performing follow-up work •science •society
- 77. http://wurmlab.github.io • Understanding/visualising/analysing/massaging big data is hard. • Biology/life is complex. • Biologists lack computational training. • Field is young. • Analysis tools (generally) suck: • badly written • badly tested • hard to install • output quality… often questionable. • Data sizes keep growing! • Data formats keep changing :( Genome bioinformatics is hardBiology is harder than (many) other data sciences
- 78. http://wurmlab.github.io Some sources of inspiration
- 80. http://wurmlab.github.io Community Page Best Practices for Scientific Computing Greg Wilson1 *, D. A. Aruliah2 , C. Titus Brown3 , Neil P. Chue Hong4 , Matt Davis5 , Richard T. Guy6¤ , Steven H. D. Haddock7 , Kathryn D. Huff8 , Ian M. Mitchell9 , Mark D. Plumbley10 , Ben Waugh11 , Ethan P. White12 , Paul Wilson13 1 Mozilla Foundation, Toronto, Ontario, Canada, 2 University of Ontario Institute of Technology, Oshawa, Ontario, Canada, 3 Michigan State University, East Lansing, Michigan, United States of America, 4 Software Sustainability Institute, Edinburgh, United Kingdom, 5 Space Telescope Science Institute, Baltimore, Maryland, United States of America, 6 University of Toronto, Toronto, Ontario, Canada, 7 Monterey Bay Aquarium Research Institute, Moss Landing, California, United States of America, 8 University of California Berkeley, Berkeley, California, United States of America, 9 University of British Columbia, Vancouver, British Columbia, Canada, 10 Queen Mary University of London, London, United Kingdom, 11 University College London, London, United Kingdom, 12 Utah State University, Logan, Utah, United States of America, 13 University of Wisconsin, Madison, Wisconsin, United States of America Introduction Scientists spend an increasing amount of time building and using software. However, most scientists are never taught how to do this efficiently. As a result, many are unaware of tools and practices that would allow them to write more reliable and maintainable code with less effort. We describe a set of best practices for scientific software development that have solid foundations in research and experience, and that improve scientists’ productivity and the reliability of their software. Software is as important to modern scientific research as telescopes and test tubes. From groups that work exclusively on computational problems, to traditional laboratory and field scientists, more and more of the daily operation of science revolves around developing new algorithms, managing and analyzing the large amounts of data that are generated in single research projects, combining disparate datasets to assess synthetic problems, and other computational tasks. Scientists typically develop their own software for these purposes because doing so requires substantial domain-specific knowledge. As a result, recent studies have found that scientists typically spend 30% or more of their time developing software [1,2]. However, 90% or more of them are primarily self-taught [1,2], and therefore lack exposure to basic software development practices such as writing maintainable code, using version control and issue error from another group’s code was not discovered until after publication [6]. As with bench experiments, not everything must be done to the most exacting standards; however, scientists need to be aware of best practices both to improve their own approaches and for reviewing computational work by others. This paper describes a set of practices that are easy to adopt and have proven effective in many research settings. Our recommenda- tions are based on several decades of collective experience both building scientific software and teaching computing to scientists [17,18], reports from many other groups [19–25], guidelines for commercial and open source software development [26,27], and on empirical studies of scientific computing [28–31] and software development in general (summarized in [32]). None of these practices will guarantee efficient, error-free software development, but used in concert they will reduce the number of errors in scientific software, make it easier to reuse, and save the authors of the software time and effort that can used for focusing on the underlying scientific questions. Our practices are summarized in Box 1; labels in the main text such as ‘‘(1a)’’ refer to items in that summary. For reasons of space, we do not discuss the equally important (but independent) issues of reproducible research, publication and citation of code and data, and open science. We do believe, however, that all of these will be much easier to implement if scientists have the skills we describe. Education A Quick Guide to Organizing Computational Biology Projects William Stafford Noble1,2 * 1 Department of Genome Sciences, School of Medicine, University of Washington, Seattle, Washington, United States of America, 2 Department of Computer Science and Engineering, University of Washington, Seattle, Washington, United States of America Introduction Most bioinformatics coursework focus- es on algorithms, with perhaps some components devoted to learning pro- gramming skills and learning how to use existing bioinformatics software. Un- fortunately, for students who are prepar- ing for a research career, this type of curriculum fails to address many of the day-to-day organizational challenges as- sociated with performing computational experiments. In practice, the principles behind organizing and documenting computational experiments are often learned on the fly, and this learning is strongly influenced by personal predilec- tions as well as by chance interactions with collaborators or colleagues. The purpose of this article is to describe one good strategy for carrying out com- putational experiments. I will not describe profound issues such as how to formulate hypotheses, design experiments, or draw conclusions. Rather, I will focus on relatively mundane issues such as organiz- ing files and directories and documenting understanding your work or who may be evaluating your research skills. Most com- monly, however, that ‘‘someone’’ is you. A few months from now, you may not remember what you were up to when you created a particular set of files, or you may not remember what conclusions you drew. You will either have to then spend time reconstructing your previous experiments or lose whatever insights you gained from those experiments. This leads to the second principle, which is actually more like a version of Murphy’s Law: Everything you do, you will probably have to do over again. Inevitably, you will discover some flaw in your initial preparation of the data being analyzed, or you will get access to new data, or you will decide that your param- eterization of a particular model was not broad enough. This means that the experiment you did last week, or even the set of experiments you’ve been work- ing on over the past month, will probably need to be redone. If you have organized and documented your work clearly, then repeating the experiment with the new under a common root directory. The exception to this rule is source code or scripts that are used in multiple projects Each such program might have a projec directory of its own. Within a given project, I use a top-leve organization that is logical, with chrono logical organization at the next level, and logical organization below that. A sample project, called msms, is shown in Figure 1 At the root of most of my projects, I have a data directory for storing fixed data sets, a results directory for tracking computa tional experiments peformed on that data a doc directory with one subdirectory per manuscript, and directories such as src for source code and bin for compiled binaries or scripts. Within the data and results directo ries, it is often tempting to apply a similar logical organization. For example, you may have two or three data sets agains which you plan to benchmark your algorithms, so you could create one directory for each of them under data In my experience, this approach is risky because the logical structure of your fina
- 82. http://wurmlab.github.io Specific Approaches/Tools 1. Write code for humans
- 83. http://wurmlab.github.io Write code for humans (not computers!) • For • yourself • colleagues / collaborators • reviewers • other random people who may reuse/improve your code • Respect conventions (e.g., a style guide)
- 84. te Damian ConwayUse whitespace/indentation! e Damian Conway Same information
- 85. Line length Strive to limit your code to 80 characters per line. This fits comfortably on a printed page with a reasonably sized font. If you find yourself running out of room, this is a good indication that you should encapsulate some of the work in a separate function. ant_measurements <- read.table(file = '~/Downloads/Web/ant_measurements.txt', header=TRUE, se ant_measurements <- read.table(file = '~/Downloads/Web/ant_measurements.txt', header = TRUE, sep = 't', col.names = c('colony', 'individual', 'headwidth', 'mass') ) ant_measurements <- read.table(file = '~/Downloads/Web/ant_measurements.txt', header=TRUE, sep='t', col.names = c('colony', 'individual', 'headwidth', 'mass')) R style guide extract http://r-pkgs.had.co.nz/style.html
- 86. R style guide extract http://r-pkgs.had.co.nz/style.html
- 88. http://wurmlab.github.io Write code for humans (not computers!) • For • yourself • colleagues / collaborators • reviewers • other random people who may want to reuse your code • Respect conventions (e.g., a style guide) • Don't optimise (generally…)
- 89. http://wurmlab.github.io Code reviews: ask a peer to (critically) read your analysis code. Or do peer-programming sessions
- 90. http://wurmlab.github.io Specific Approaches/Tools 1. Write code for humans 2. Organise mindfully
- 91. Eliminate redundancy DRY: Don’t RepeatYourself & don't reinvent the wheel. Organise mindfully
- 92. http://wurmlab.github.io Organise mindfully http://bit.ly/projectstruct Choose a standard/ template and stick to it! Choose a standard/ template and stick to it!
- 94. http://wurmlab.github.io Specific Approaches/Tools 1. Write code for humans 2. Organise mindfully 3. Plan for mistakes
- 95. Automatically check consistency with style guide install.packages("lint") # once library(lint) # everytime lint("file_to_check.R")
- 96. http://wurmlab.github.io Create code tests that are easy to run • Unit tests == checking edge cases to see if the function works # do your stuff # e.g. define speed() function library(testthat) expect_that(speed(km = 0, minutes = 60), equals(0)) expect_that(speed(km = 60, minutes = 60), equals(1)) expect_that(speed(km = -4, minutes = 60), throws_error()) expect_that(nrow(significant_SNPs), 42) expect_that(my_model, is_a("lm")) • Integration tests == "full analysis" but on small data with known results • e.g. on fakeVCF genotype file of 2 loci (one true positive, one true negative) • Add sanity checks. E.g. the following should fail rather than return something incorrect. speed(km= "twenty", minutes=20) speed(km = -4, minutes = 60)
- 97. http://wurmlab.github.io "Continuous integration": Tests should run automagically. So you don't have to remember (or find time) to do it. "http://github.org Tests run automatically http://travis-ci.org If unexpected result: #
- 98. http://wurmlab.github.io Specific Approaches/Tools 1. Write code for humans 2. Organise mindfully 3. Plan for mistakes 4. Use tools that reduce risks
- 99. http://wurmlab.github.io Use tools that reduce risks • Ensure computers are set up for productivity. E.g.,: • use GNU parallel on an 80-core machine when more appropriate than submitting to queue • If you need to make a "pipeline", use software designed for this. E.g.: • Snakemake • Nextflow • (etc) • too many examples to discuss here
- 100. knitr/rmarkdown/ jupyter Analysis & report in one. analysis.Rmd A minimal R Markdown example I know the value of pi is 3.1416, and 2 times pi is 6.2832. To c library(knitr); knit( minimal.Rmd ) A paragraph here. A code chunk below: 1+1 ## [1] 2 .4-.7+.3 # what? it is not zero! ## [1] 5.551e-17 Graphics work too library(ggplot2) qplot(speed, dist, data = cars) + geom_smooth() ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●●● ● ● 0 40 80 120 5 10 15 20 speed dist Figure 1: A scatterplot of cars
- 102. How to get users to adopt good practices? • Carrot (dual-benefit): • Use their motivation to have an easier life. "their motivation is the database" "they see it, they understand it" -Thomasz? on SEEK • Piggyback off that so they do things better ("by stealth" -Carol) • Stick: • When you're reviewing publications/grants • Politics: • Encourage funders / journals to require good practices.
- 104. Summary • Ants are cool • Biology is hard • We need to handle data better
- 105. y.wurm@qmul.ac.uk @yannick__ https://wurmlab.github.io @ Queen Mary U London Rodrigo Pracana Anurag Priyam @yeban Eckart Stolle Bruno Vieira @bmpvieira R Nichols & sbcsEvolve R Christie & T King / ITSR Apocrita Laurent Keller lab @ Lausanne J Wang, D Shoemaker,O Riba-Grognuz, M Nipitwattanaphon Ioannis Xenarios @ SIB DeWayne Shoemaker @ USDA Thanks!