If you like this presentation – show it...
Crowdsourcing Biology: The Gene Wiki, BioGPS and GeneGames.org May 14, 2014 CBIIT Slides: slideshare.net/andrewsu Citizen Science!
Few genes are well annotated… 2 Data: NCBI, February 2013
… because the literature is sparsely curated? 3
… because the literature is sparsely curated? 4
5 311,696 articles (1.5% of PubMed) have been cited by GO annotations
6 0 Sooner or later, the research community will need to be involved in the annotation effort to scale up to the rate of data generation.
The Long Tail is a prolific source of content 7 News : Video: Product reviews: Food reviews: Talent judging: Newspapers TV/Hollywood Consumer reports Food critics Olympics Blogs YouTube Amazon reviews Yelp American Idol
Wikipedia is reasonably accurate 8
Wikipedia has breadth and depth 9 http://en.wikipedia.org/wiki/Wikipedia:Size_comparisons, July 2008
10 We can harness the Long Tail of scientists to directly participate in the gene annotation process.
From crowdsourcing to structured data 11
Filtering, extracting, and summarizing PubMed Documents Concepts Review article
Filtering, extracting, and summarizing PubMed Documents Concepts
Wiki success depends on a positive feedback 14 Gene wiki page utility Number of users Number of contributors 100 1 200 2
10,000 gene “stubs” within Wikipedia 15 Protein structure Symbols and identifiers Tissue expression pattern Gene Ontology annotations Links to structured databases Gene summary Protein interactions Linked references Huss, PLoS Biol, 2008
Gene Wiki has a critical mass of readers 16 Total: 4.0 million views / month Huss, PLoS Biol, 2008; Good, NAR, 2011
Gene Wiki has a critical mass of editors 17 Increase of ~10,000 words / month from >1,000 edits Currently 1.42 million words Approximately equal to 230 full-length articles Good, NAR, 2011
A review article for every gene is powerful 18 References to the literature Hyperlinks to related concepts Reelin: 98 editors, 703 edits since July 2002 Heparin: 358 editors, 654 edits since June 2003 AMPK: 109 editors, 203 edits since March 2004 RNAi: 394 editors, 994 edits since October 2002
Making the Gene Wiki more computable 19 Structured annotations Free text
Filling the gaps in gene annotation 20 6319 novel GO annotations 2147 novel DO annotations
Gene Wiki content improves enrichment analysis 21 GO term Gene list Concept recognition PubMed abstracts Enrichment analysis GO:0007411 axon guidance (GO:0007411) 264 genes Linked genes through PubMed P = 1.55 E-20 811 articles
Gene Wiki content improves enrichment analysis 22 GO term Gene list Concept recognition PubMed abstracts Enrichment analysis GO:0006936 GO:0006936 muscle contraction (GO:0006936) 87 genes Linked genes through PubMed Linked genes through PubMed + Gene Wiki P = 1.0 P = 1.22 E-09 251 articles 87 articles
Gene Wiki content improves enrichment analysis 23 p-value (PubMed only) p-value (PubMed + GW) Muscle contraction More significant PubMed + GW More significant PubMed only Good BM et al., BMC Genomics, 2011
Making the Gene Wiki more computable 24 Structured annotations Free text Analyses
Making the Gene Wiki more computable 25 Structured annotations Free text Databases
Expansion through outreach and incentives 26
Cardiovascular Gene Wiki Portal 27 CAMK2D -- CaM kinase II subunit delta CSRP3 -- Cysteine and glycine-rich protein 3 GJA1 -- Gap junction alpha-1 protein / Connexin-43 MAPK14 -- Mitogen-activated protein kinase 14 / p38-? MYL7 -- Myosin regulatory light chain 2, atrial isoform MYL2 -- Myosin regulatory light chain 2, ventricular/cardiac isoform PECAM1 -- Platelet endothelial cell adhesion molecule/CD31 RYR2 -- Ryanodine receptor 2 ATP2A2 -- Sarcoplasmic/endoplasmic reticulum calcium ATPase 2 / SERCA2 TNNI3 -- Troponin I, cardiac muscle TNNT2 -- Troponin T, cardiac muscle Peipei Ping UCLA
The Long Tail of scientists is a valuable source of information on gene function 28
From crowdsourcing to structured data 29
Gene databases are numerous and overlapping 30 … and hundreds more …
Why is there so much redundancy? 31 Users Requests Resources Time Community development BioGPS emphasizes community extensibility
Why do developers define the gene report view? 32 BioGPS emphasizes user customizability
Community extensibility and user customizability 33
Utility: A simple and universal plugin interface 34
Utility: A simple and universal plugin interface 35
Utility: A simple and universal plugin interface 36
Utility: A simple and universal plugin interface 37
Utility: A simple and universal plugin interface 38
Utility: A simple and universal plugin interface 39 Total of > 540 gene-centric online databases registered as BioGPS plugins
Users: BioGPS has critical mass 40 Daily pageviews
Contributors: Explicit and implicit knowledge 41 540 plugins registered (>300 publicly shared) by over 120 users spanning 280+ domains
Gene Annotation Query as a Service 42 http://mygene.info High performance 3M hits/month Highly scalable 13k species 16M genes Weekly data updates JSON output REST interface Python/R/JS libraries
The Long Tail of bioinformaticians can collaboratively build a gene portal. 43
From crowdsourcing to structured data 44
The biomedical literature is growing fast 45
Information Extraction 46 Find mentions of high level concepts in text Map mentions to specific terms in ontologies Identify relationships between concepts
Disease mentions in PubMed abstracts 47 NCBI Disease corpus 793 PubMed abstracts (100 development, 593 training, 100 test) 12 expert annotators (2 annotate each abstract) 6,900 “disease” mentions Dogan, Rezarta, and Zhiyong Lu. "An improved corpus of disease mentions in PubMed citations." Proceedings of the 2012 Workshop on Biomedical Natural Language Processing. Association for Computational Linguistics.
Four types of disease mentions 48 Specific Disease: “Diastrophic dysplasia” Disease Class: “Cancers” Composite Mention: “prostatic , skin , and lung cancer” Modifier: ..the “familial breast cancer” gene , BRCA2.. Dogan, Rezarta, and Zhiyong Lu. "An improved corpus of disease mentions in PubMed citations." Proceedings of the 2012 Workshop on Biomedical Natural Language Processing. Association for Computational Linguistics.
Question: Can a group of non-scientists collectively perform concept recognition in biomedical texts? 49
The Turk 50 http://en.wikipedia.org/wiki/The_Turk
The Turk 51 http://en.wikipedia.org/wiki/The_Turk
Amazon Mechanical Turk (AMT) 52 For each task, specify: a qualification test how many workers per task how much we will pay per task Manages: parallel execution of jobs worker access to tasks via qualification tests payments task advertising 1. Create tasks 2. Execute 3. Aggregate
Instructions to workers 53 Highlight all diseases and disease abbreviations “...are associated with Huntington disease ( HD )... HD patients received...” “The Wiskott-Aldrich syndrome ( WAS ) , an X-linked immunodeficiency…” Highlight the longest span of text specific to a disease “... contains the insulin-dependent diabetes mellitus locus …” Highlight disease conjunctions as single, long spans. “... a significant fraction of familial breast and ovarian cancer , but undergoes…” Highlight symptoms - physical results of having a disease “XFE progeroid syndrome can cause dwarfism, cachexia, and microcephaly. Patients often display learning disabilities, hearing loss, and visual impairment.
Qualification test 54 Test #1: “Myotonic dystrophy ( DM ) is associated with a ( CTG ) in trinucleotide repeat expansion in the 3-untranslated region of a protein kinase-encoding gene , DMPK , which maps to chromosome 19q13 . 3 . ” Test #2: “Germline mutations in BRCA1 are responsible for most cases of inherited breast and ovarian cancer . However , the function of the BRCA1 protein has remained elusive . As a regulated secretory protein , BRCA1 appears to function by a mechanism not previously described for tumour suppressor gene products.” Test #3: “We report about Dr . Kniest , who first described the condition in 1952 , and his patient , who , at the age of 50 years is severely handicapped with short stature , restricted joint mobility , and blindness but is mentally alert and leads an active life . This is in accordance with molecular findings in other patients with Kniest dysplasia and…” 26 yes / no questions
Qualification test results 55
Simple annotation interface 56 Click to see instructions Highlight disease mentions
Experimental design Task: Identify the disease mentions in the 593 abstracts from the NCBI disease corpus $0.06 per Human Intelligence Task (HIT) HIT = annotate one abstract from PubMed 5 workers annotate each abstract 57
Aggregation function based on simple voting 58 58 1 or more votes (K=1) K=2 K=3 K=4
Comparison to gold standard 59 593 documents 7 days 17 workers $192.90
Comparison to gold standard 60 Max F = 0.69 0.79 0.82 k=1 2 3 2 3 4 5 0.85 k=1 N = 3 6 9 12 15 18 7 8 0.85 0.85
Comparison to gold standard 61 Max F = 0.69 0.79 0.82 k=1 2 3 2 3 4 5 0.85 k=1 N = 3 6 9 12 15 18 7 8 0.85 0.85
Comparison to gold standard 62 Max F = 0.69 0.79 0.82 k=1 2 3 2 3 4 5 0.85 k=1 N = 3 6 9 12 15 18 7 8 0.85 0.85
Comparison to gold standard 63 Max F = 0.69 0.79 0.82 k=1 2 3 2 3 4 5 0.85 k=1 N = 3 6 9 12 15 18 7 8 0.85 0.85
Comparisons to text-mining algorithms 64
Comparisons to human annotators 65 Average level of agreement between expert annotators (stage 1) F = 0.76
Comparisons to human annotators 66 F = 0.76 F = 0.87 Average level of agreement between expert annotators (stage 2)
67 In aggregate, our worker ensemble is faster, cheaper and as accurate as a single expert annotator for disease concept recognition.
Information Extraction 68 Find mentions of high level concepts in text Map mentions to specific terms in ontologies Identify relationships between concepts
Annotating the relationships 69 This molecule inhibits the growth of a broad panel of cancer cell lines, and is particularly efficacious in leukemia cells, including orthotopic leukemia preclinical models as well as in ex vivo acute myeloid leukemia (AML) and chronic lymphocytic leukemia (CLL) patient tumor samples. Thus, inhibition of CDK9 may represent an interesting approach as a cancer therapeutic target especially in hematologic malignancies. therapeutic target subject predicate object GENE DISEASE
Citizen Science at Mark2Cure.org 70
The Long Tail of citizen scientists can collaboratively annotate biomedical text. 71
72 Doug Howe, ZFIN John Hogenesch, U Penn Jon Huss, GNF Luca de Alfaro, UCSC Angel Pizzaro, U Penn Faramarz Valafar, SDSU Pierre Lindenbaum, Fondation Jean Dausset Michael Martone, Rush Konrad Koehler, Karo Bio Warren Kibbe, Simon Lim, Northwestern Lynn Schriml, U Maryland Paul Pavlidis, U British Columbia Peipei Ping, UCLA Many Wikipedia editors WP:MCB Project Collaborators Contact http://sulab.org email@example.com @andrewsu +Andrew Su Citizen Science logo based on http://thenounproject.com/term/teamwork/39543/
Related AMT work 73  Zhai et al 2013, used similar protocol to tag medication names in clinical trials descriptions. F = 0.88 compared to gold standard  Burger et al, using microtask workers to identify relationships between genes and mutations.  Aroyo & Welty, used workers to identify relations between concepts in medical text.  Zhai H. et al (2013) ”Web 2.0-Based Crowdsourcing for High-Quality Gold Standard Development in Clinical Natural Language Processing” J Med Internet Res  Burger, John, et al. (2014) "Hybrid curation of gene-mutation relations combining automated extraction and crowdsourcing.” Mitre technical report  Aroyo, Lora, and Chris Welty. Harnessing disagreement in crowdsourcing a relation extraction gold standard. Tech. Rep. RC25371 (WAT1304-058), IBM Research, 2013.