BioNLP Tutorial PSB 2006 Wailea, Maui, HI K. Bretonnel Cohen Olivier Bodenreider Lynette Hirschman

The Biological Data Cycle MEDLINE Literature Collections Experimental Data Ontologies Expert Curation Databases SwissProt Genbank Bottleneck: getting knowledge from literature to databases Solution: text mining 1

MEDLINE 1. Select papers 2. List genes for curation 3. Curate genes from paper Model Organism Curation Pipeline 1

Double exponential growth in the literature New entries in Medline with publication date in Jan-Aug 2005: 431,478 (avg. 1775/ day) 1

Examples of BioNLP in action 1

1

1

Application types Information retrieval: find documents in response to an “information need” p53 Resistance to apoptosis, increased growth potential, and altered gene expression in cells that survived genotoxic hexavalent chromium exposure. PMID:

Application types Question-answering: question as input, answer as output What is BRCA1 ? A gene located on the seventeenth chromosome associated with a risk of breast and ovarian cancer 2 (Yu and Sable 2005)

Application types Summarization –Input: one or more texts –Output: single (shorter) text Information extraction: Information extraction systems find statements about some specified type of relationship in text. Entity identification is a necessary prerequisite to information extraction. Information retrieval: Information retrieval is classically defined as the location of documents that are relevant to some information need. PubMed is a premier example of a sophisticated biomedical information retrieval system. Summarization systems benefit from high-performance entity identification and normalization. Other approaches involve information extraction. 2 Ling et al. (multiple documents) Lu et al. (single document)

Application types Information extraction: relationships between things BINDING_EVENT Binder: Bound: 2

Application types Met28 binds to DNA. BINDING_EVENT Binder: Met28 Bound: DNA 2 Lussier (gene/phenotype) Maguitman (protein/family) Chun (gene/disease) Höglund (protein/location) Stoica (protein/function)

Application types HSP60 Hsp-60 heat shock protein 60 Cerberus wingless Ken and Barbie the Entity identification 3

Application types Entity normalization: find concepts in text and map them to unique identifiers A locus has been found, an allele of which causes a modification of some allozymes of the enzyme esterase 6 in Drosophila melanogaster. There are two alleles of this locus, one of which is dominant to the other and results in increased electrophoretic mobility of affected allozymes. The locus responsible has been mapped to on the standard genetic map (Est-6 is at ). Of 13 other enzyme systems analyzed, only leucine aminopeptidase is affected by the modifier locus. Neuraminidase incubations of homogenates altered the electrophoretic mobility of esterase 6 allozymes, but the mobility differences found are not large enough to conclude that esterase 6 is sialylated. 3

Perfect entity identification finds 5 mentions; they correspond to just 2 genes: –FBgn (esterase 6) –FBgn (leucine aminopeptidase) A locus has been found, an allele of which causes a modification of some allozymes of the enzyme esterase 6 in Drosophila melanogaster. There are two alleles of this locus, one of which is dominant to the other and results in increased electrophoretic mobility of affected allozymes. The locus responsible has been mapped to on the standard genetic map (Est-6 is at ). Of 13 other enzyme systems analyzed, only leucine aminopeptidase is affected by the modifier locus. Neuraminidase incubations of homogenates altered the electrophoretic mobility of esterase 6 allozymes, but the mobility differences found are not large enough to conclude that esterase 6 is sialylated. Application types 3

Partial list of synonyms for FBgn : –Esterase 6 –Carboxyl ester hydrolase –CG6917 –Est6 –Est-D –Est-5 3 Chun (gene/disease) Johnson (ontology alignment) Stoica (gene/function) Vlachos (FlyBase mapping)

Biological Nomenclature: “V-SNARE” SNAP Receptor Vesicle SNARE V-SNARE N-Ethylmaleimide-Sensitive Fusion Protein Soluble NSF Attachment Protein Maleic acid N-ethylimide Vesicle Soluble Maleic acid N-ethylimide Sensitive Fusion Protein Attachment Protein Receptor (A. Morgan) 4

The Biological Data Cycle MEDLINE Literature Collections Experimental Data Ontologies Expert Curation Databases SwissProt Genbank What’s the organizing principle for all of this? 4

Organizing principles Biomedical literature Biomedical literature MeSH Genome annotations Genome annotations GO Model organisms Model organisms NCBI Taxonomy Genetic knowledge bases OMIM Clinical repositories Clinical repositories SNOMED Other subdomains Other subdomains … Anatomy UWDA UMLS 4

Organizing principles 4

Neurofibromatosis type 2 (NF2) is often not recognised as a distinct entity from peripheral neurofibromatosis. NF2 is a predominantly intracranial condition whose hallmark is bilateral vestibular schwannomas. NF2 results from a mutation in the gene named merlin, located on chromosome 22. (Uppal, S., and A. P. Coatesworth. “Neurofibromatosis Type 2.” Int J Clin Pract, 57, no. 8, 2003, pp ) Ontologies as text mining resources 4

Neurofibromatosis type 2 (NF2) is often not recognised as a distinct entity from peripheral neurofibromatosis. NF2 is a predominantly intracranial condition whose hallmark is bilateral vestibular schwannomas. NF2 results from a mutation in the gene named merlin, located on chromosome 22. Ontologies as text mining resources Disease Tumor Gene Chromosome vestibular schwannoma manifestation of neurofibromatosis 2 neurofibromatosis 2 associated with mutation of merlin merlin located on chromosome 22 Tumor manifestation of Disease Disease associated with mutation of Gene Gene located on Chromosome 4

What’s the state of the art? Tasks differ greatly: finding human protein interactions (Bunescu ‘05) may be harder than finding “inhibition” relations (Pustejovsky ‘02) Need a CASP-style competitive evaluation Precision ≈ Specificity Recall ≈ Sensitivity 4

What’s the state of the art? KDD Cup (2002) TREC Genomics (2003, 2004, 2005) BioCreAtIvE (2004) BioNLP (2004)

MEDLINE 1. Select papers KDD 2002, TREC Genomics List genes for curation BioCreAtIvE entity identification and entity normalization tasks 3. Curate genes from paper BioCreAtIvE information extraction task: PDB → Gene Ontology What’s the state of the art? 5

**F-measure is balanced precision and recall: 2*P*R/(P+R) Recall: # correctly identified/# possible correct Precision:# correctly identified/# identified Yeast results good: High: 0.93 F Smallest vocab Short names Little ambiguity Fly: 0.82 F High ambiguity Mouse: 0.79 F Large vocabulary Long names What’s the state of the art? 3

Blaschke et al. 5

What’s the state of the art? Cellular Component: 34.61% (561/1621) Molecular Function: 33.00%(933/2827) Biological Process: 23.02% (1011/4391) Cellular component is easier because task is relation between “entities” located_in (protein,cell_component) Biological process is hardest because it is the most abstract Blaschke et al. 5

2.5 types of solutions Rule-based –Patterns –Grammars Statistical/machine learning –Labelled training data –Noisy training data Hybrid statistical/rule-based Höglund (information extraction, gene → localiz.) Maguitman (info. extract., SWISSPROT → Pfam) Vlachos (entity normalization, gene → FlyBase) Stoica (gene → GO code) Chun (IE, multiple gene -> UMLS disease) Ling (summarization, FlyBase) Johnson (ontology alignment, GO → other OBO) Lu (summarization, Entrez Gene → GeneRIFs) Lussier (info. extraction, GOA -> phenotype) Vlachos (coreference, FlyBase & Sequence Ont.) 5

Common tools/techniques “Stop word” removal: eliminate features that are rarely helpful the, a, and… (Porter) stemming: convert inflected words to their roots promot, mitochondri, cytochrom POS: “part of speech”— ≈80 categories 5

Why text mining is difficult Variability Pervasive ambiguity at every level of analysis 5

Why text mining is difficult Met28 binds to DNA …binding of Met28 to DNA… …Met28 and DNA bind… …binding between Met28 and DNA… …Met28 is sufficient to bind DNA… …DNA bound by Met28… 2(6)

Why text mining is difficult …binding of Met28 to DNA… …binding under unspecified conditions of Met28 to DNA… …binding of this translational variant of Met28 to DNA… …binding of Met28 to upstream regions of DNA… 2(6)

Why text mining is difficult …binding under unspecified conditions of this translational variant of Met28 to upstream regions of DNA… 3(6)

Why text mining is difficult Document segmentation Sentence segmentation Tokenization Part of speech tagging Parsing 5

Why text mining is difficult Here, we show that Bifocal (Bif), a putative cytoskeletal regulator, is a component of the Msn pathway for regulating R cell growth targeting. bif displays strong genetic interaction with msn. (Ruan et al. 2002) F-measure MaxEnt_1.40 MaxEnt_2.67 KeX.95 LingPipe.96 (Baumgartner, in prep.) 6

Why text mining is difficult lead 69 tokens in GENIA –“bare stem” verb: 34 –3 rd person singular present tense verb: 29 –Noun: 3 –Past tense verb: 2 –Past participle: 1 6

Why text mining is difficult HUNK Human natural killer (cell type) HUN kinase (gene/protein) Radiological/orthopedic classification scheme Piece of something 6

NaCT is expressed in liver, testis and brain in rat and shows preference for citrate over dicarboxylates… (GeneRIF : ) NACT: neoadjuvant chemotherapy (PMID ) N-acetyltransferase (PMID ) Na+-coupled citrate transporter (PMID ) Why text mining is difficult 6

NaCT is expressed in liver, testis and brain in rat and shows preference for citrate over dicarboxylates… (GeneRIF : ) (liver), (testis) and (brain in rat) liver, (testis and brain in rat) (liver, testis and brain in rat) 6

Why text mining is difficult NaCT is expressed in liver, testis and brain in rat and shows preference for citrate over dicarboxylates… (GeneRIF : ) shows preference for (citrate over dicarboxylates) shows preference (for citrate) (over dicarboxylates) 7

Why text mining is difficult regulation of cell migration and proliferation (PMID …) serine phosphorylation, translocation, and degradation of IRS-1 (PMID ) !proliferation and regulation of cell migration !regulation of proliferation and cell migration regulation of cell migration and regulation of cell proliferation 7

Why text mining is difficult regulation of cell migration and proliferation (PMID …) serine phosphorylation, translocation, and degradation of IRS-1 (PMID ) !degradation of IRS-1, translocation, and serine phosphorylation !serine phosphorylation, serine translocation, and serine degradation (of IRS-1) 7

Most biomedical text mining to date: “ungrounded” Drosophila OBP76a is necessary for fruit flies to respond to the aggregation pheromone 11-cis vaccenyl acetate (PMID ) lush is completely devoid of evoked activity to the pheromone 11-cis vaccenyl acetate (VA), revealing that this binding protein is absolutely required for activation of pheromone-sensitive chemosensory neurons (PMID ) 7 Entrez Gene ID:40136

The next step Text mining can be key tool for linking biological knowledge from the literature to structured data in biological databases… …and databases to each other. 7

Papers in the text mining session 5 papers on linkage to ontologies Höglund et al.: generating cellular localization annotations Lussier et al.: PhenoGO for capture of phenome data Stoica and Hearst: functional annotation of proteins Johnson et al.: ontology alignments Vlachos et al.: ontology for name extraction, anaphora 2 papers linking other sets of resources Maguitman et al. on “bibliome” to reproduce Pfam classes Chun et al. on linking genes and diseases 2 papers on summarization, using linked resources Lu et al.: automated GeneRIF extraction Ling et al.: automated gene summary generation 7

Acknowledgements Alex Morgan for several slides Christian Blaschke for data and slides Bill Baumgartner for sentence segmenter performance data Helen Johnson for data on POS ambiguity in GENIA Lu Zhiyong for syntactic ambiguity examples Larry Hunter for current PubMed graph 7

How big is a humuhumunukunukuapua’a?