Finding Functional Gene Relationships Using the Semantic Gene Organizer (SGO) Kevin Heinrich Master’s Defense July 16, 2004

Outline Problem / Goals Related Work Information Retrieval –Vector Space Model –Latent Semantic Indexing (LSI) Biological Databases SGO Use & Results

Problem Biological tools are creating vast amounts of data. Current techniques are time-consuming and expensive. Want to know phenotype (function) from genotype (structure/sequence).

Goals Develop a tool to aid researchers in finding and understanding functional gene relationships. Use information that covers whole genome, e.g. literature.

Related Work Jenssen et al. (2001) developed PubGene. –Literature network –Assigns functional association if there is a co- occurrence of gene symbols Wilkinson and Huberman (2004) expanded this idea to find communities of related genes. Yandell and Majoros (2002) use natural language processing techniques to identify nature of relationships.

Related Work Most all literature-based techniques rely on term co-occurrence. What about gene aliases? Solution: Apply a more robust technique.

Information Retrieval Vector Space Model Documents are parsed into tokens. Tokens are assigned a weight of, w ij, of i th token in j th document. An m x n term-by-document matrix, A, is created where –Documents are m-dimensional vectors. –Tokens are n-dimensional vectors.

Information Retrieval Term Weights Term weights are the product of a local and global component tf idf idf2

Information Retrieval Term Weights (cont’d) log-entropy Goal is to give distinguishing terms more weight.

Information Retrieval Query & Similarity Queries are represented by a pseudo-document vector Similarity is the cosine of the angle between document vectors.

Information Retrieval Latent Semantic Indexing (LSI) LSI performs a truncated SVD on A = UΣV T U is the m x n matrix of eigenvectors of AA T V T is the r x n matrix of eigenvectors of A T A Σ is the r x r diagonal matrix containing the r nonnegative singular values of A r is the rank of A A rank-k approximation is given by A k = U k Σ k V k T

Information Retrieval LSI (cont’d) Document-to-document similarity is Queries are projected into low-rank approximation space

Information Retrieval LSI (cont’d) Scaled document vectors can be computed once and stored for quick retrieval. The lower-dimensional space forces queries and documents to be compared in a more conceptual manner and saves storage. Choice of number of factors is an open question. End Effect: LSI can find similarities between documents that have no term co-occurrence.

Information Retrieval Evaluation Measures Precision – ratio of relevant returned documents to the total number of returned documents. Recall – ratio of relevant returned documents to the total number of relevant documents. Goal is to have high precision at all levels of recall. Systems are often evaluated by average precision (AP), which is the average of 11 interpolated precision values at the decile ranges.

Biological Databases MEDLINE MEDLINE (NLM) –Contains 14+ million references to journal articles with a concentration in medicine –Span over 4,600 journals worldwide –1966 to present –~500,000 citations added annually –Each citation is manually indexed with MeSH terms.

Biological Databases PubMed PubMed –Retrieves articles from MEDLINE and other journals. –Can be queried via any combination of attributes.

Biological Databases LocusLink NCBI human-curated database Single query interface to a comprehensive directory for genes and gene reference sequences for key genomes. Provides links to related records in PubMed and other citations when applicable. Provides RefSeq Summary of gene function and links to key MEDLINE citations relevant to each gene.

Biological Databases Overview MEDLINE has lots information –Not all articles relate to genes –Gene terminology problem LocusLink does not cover all relevant citations, but a representative few.

Biological Databases Gene Document Construction Concatenate titles and abstracts of MEDLINE citations cross-referenced in Human, Rat, and Mouse LocusLink entries. Sequencing abstracts included – noise LocusLink references are not comprehensive, so recall of all relevant abstracts is not guaranteed.

SGO Primarily uses LSI to rank genes. Enables user to specify query method –Gene query –Keyword query –Number of factors –Show latent matches Saves previous query sessions.

SGO Interface

SGO Interface (cont’d)

SGO Trees Unfortunately, ranked lists mean little to biologists. Pairwise distances can be formed into a matrix where is the similarity between documents i and j

SGO Trees (cont’d) Fitch-Margoliash (1967) method in PHYLIP is applied to D to generate hierarchical trees. Thresholds can be applied to self-similarity matrix to produce graphs.

SGO Hierarchical Tree

SGO Graph or Nodal Tree

SGO Coding Issues Web interface – must be interactive –Queries are processed on click –Document collections are parsed offline –Trees are constructed offline Storage will eventually become an issue.

Results Test Data Set 50 gene test data set was constructed. –Alzheimer’s Disease –Cancer –Development Reelin signaling pathway used as basis for evaluation –5 primary genes (directly associated) –7 secondary genes (indirectly associated)

Results Primary AP AP for 5 primary genes –61% for 5 factors –84% for 25 factors –84% for 50 factors

Results Secondary AP AP for 12 secondary genes –53% for 5 factors –59% for 25 factors –61% for 50 factors

Results Comparison LSI comparable to tf-idf for 5 primary genes Far superior to tf-idf for 12 second genes –PubMed co-citation identifies 2 of the 7 indirectly related genes –Abstract overlap of LocusLink citations fails to identify any indirectly related genes tf-idf fails on many keyword queries Tested on Gene Ontology classifications (not shown) –Similar tendencies are observed

Results Abstract Representation To simulate scaling up, decrease representation of reelin-related genes AP of 47% on 20,856 Human LocusLink abstracts

Results Hierarchical Tree

Conclusions SGO allows genes to be compared to each other and to keyword (function). SGO identifies latent relationships with promising accuracy. SGO is not meant to replace existing technologies, but to assist researchers –Verify current results –Direct future exploration

Future Work Scale up to entire genome Document construction Incorporate structural or other information for multi-modal similarity Test other models e.g. NMF, QR, etc. Interactive tree building Keep collections current