Integrating text mining into the MGI biocuration workflow.

A major challenge for functional and comparative genomics resource development is the extraction of data from the biomedical literature. Although text mining for biological data is an active research field, few applications have been integrated into production literature curation systems such as those of the model organism databases (MODs). Not only are most available biological natural language (bioNLP) and information retrieval and extraction solutions difficult to adapt to existing MOD curation workflows, but many also have high error rates or are unable to process documents available in those formats preferred by scientific journals.In September 2008, Mouse Genome Informatics (MGI) at The Jackson Laboratory initiated a search for dictionary-based text mining tools that we could integrate into our biocuration workflow. MGI has rigorous document triage and annotation procedures designed to identify appropriate articles about mouse genetics and genome biology. We currently screen approximately 1000 journal articles a month for Gene Ontology terms, gene mapping, gene expression, phenotype data and other key biological information. Although we do not foresee that curation tasks will ever be fully automated, we are eager to implement named entity recognition (NER) tools for gene tagging that can help streamline our curation workflow and simplify gene indexing tasks within the MGI system. Gene indexing is an MGI-specific curation function that involves identifying which mouse genes are being studied in an article, then associating the appropriate gene symbols with the article reference number in the MGI database.Here, we discuss our search process, performance metrics and success criteria, and how we identified a short list of potential text mining tools for further evaluation. We provide an overview of our pilot projects with NCBO's Open Biomedical Annotator and Fraunhofer SCAI's ProMiner. In doing so, we prove the potential for the further incorporation of semi-automated processes into the curation of the biomedical literature.

The Mouse Genome Database (MGD): from genes to mice--a community resource for mouse biology.

The Mouse Genome Database (MGD) forms the core of the Mouse Genome Informatics (MGI) system (http://www.informatics.jax.org), a model organism database resource for the laboratory mouse. MGD provides essential integration of experimental knowledge for the mouse system with information annotated from both literature and online sources. MGD curates and presents consensus and experimental data representations of genotype (sequence) through phenotype information, including highly detailed reports about genes and gene products. Primary foci of integration are through representations of relationships among genes, sequences and phenotypes. MGD collaborates with other bioinformatics groups to curate a definitive set of information about the laboratory mouse and to build and implement the data and semantic standards that are essential for comparative genome analysis. Recent improvements in MGD discussed here include the enhancement of phenotype resources, the re-development of the International Mouse Strain Resource, IMSR, the update of mammalian orthology datasets and the electronic publication of classic books in mouse genetics.

Signaling through the interaction of membrane-restricted stem cell factor and c-kit receptor tyrosine kinase: genetic evidence for a differential role in erythropoiesis.

Mutations of the receptor tyrosine kinase c-kit or its ligand stem cell factor (SCF), which is encoded as a soluble and membrane-associated protein by the Steel gene in mice, lead to deficiencies of germ cells, melanocytes, and hematopoiesis, including the erythroid lineage. In the present study, we have used genetic methods to study the role of membrane or soluble presentation of SCF in hematopoiesis. Bone marrow-derived stromal cells expressing only a membrane-restricted (MR) isoform of SCF induced an elevated and sustained tyrosine phosphorylation of both c-kit and erythropoietin receptor (EPO-R) and significantly greater proliferation of an erythrocytic progenitor cell line compared with stromal cells expressing soluble SCF. Transgene expression of MR-SCF in Steel-dickie (Sld) mutants resulted in a significant improvement in the production of red blood cells, bone marrow hypoplasia, and runting. In contrast, overexpression of the full-length soluble form of SCF transgene had no effect on either red blood cell production or runting but corrected the myeloid progenitor cell deficiency seen in these mutants. These data provide the first evidence of differential functions of SCF isoforms in vivo and suggest an abnormal signaling mechanism as the cause of the severe anemia seen in mutants of the Sl gene.

Overexpression of human stem cell factor impairs melanocyte, mast cell, and thymocyte development: a role for receptor tyrosine kinase-mediated mitogen activated protein kinase activation in cell differentiation.

Stem cell factor (SCF) is synthesized as both soluble (S) and membrane-associated (MA) proteins. Indirect insight into the function of MA and S isoforms of SCF has come from studies performed in Steel (Sl) mutant mice. However, the physiologic role(s) of these two isoforms remain unknown. In an attempt to better understand the in vivo role of c-kit/SCF interactions on various cell lineages, transgenic mice were generated that overexpress MA isoform of human SCF (hSCF). In murine cells, hSCF behaves as an antagonist to normal SCF function, due to interference with the interaction between endogenous murine SCF and its receptor, c-kit, encoded by the dominant white spotting (W) gene. Mice expressing the hSCF transgene display a variety of phenotypic abnormalities, which are accentuated when combined with W alleles. Here we show that mice homozygous for the hSCF transgene demonstrate a coat color deficiency seen in some mice homozygous for mild W alleles. Specifically, homozygous hSCF transgenic mice (hSCF220) display a pronounced forehead blaze, with additional white spots over the cervical region, as well as a very large belly spot. Doubly heterozygous animals that carry both a mutated W allele and the hSCF transgene also display an unusual pigment defect and a dramatic reduction in the number of dermal mast cells. Furthermore, overexpression of MA hSCF in the thymus results in abnormal thymocyte differentiation and proliferation, which is associated with reduced mitogen activated protein (MAP) kinase activation. Thus, MAP kinase activation by a receptor tyrosine kinase, such as c-kit, may be critical for the differentiation of thymocytes in vivo.

Tissue, developmental, and tumor-specific expression of divergent transcripts in Wilms tumor.

The Wilms tumor locus on chromosome 11p13 has been mapped to a region defined by overlapping, tumor-specific deletions. Complementary DNA clones representing transcripts of 2.5 (WIT-1) and 3.5 kb (WIT-2) mapping to this region were isolated from a kidney complementary DNA library. Expression of WIT-1 and WIT-2 was restricted to kidney and spleen. RNase protection revealed divergent transcription of WIT-1 and WIT-2, originating from a DNA region of less than 600 bp. Both transcripts were present at high concentrations in fetal kidney and at much reduced amounts in 5-year-old and adult kidneys. Eleven of 12 Wilms tumors classified as histopathologically heterogeneous exhibited absent or reduced expression of WIT-2, whereas only 4 of 14 histopathologically homogeneous tumors showed reduced expression. These data demonstrate a molecular basis for the pathogenetic heterogeneity in Wilms tumorigenesis.