Abstract

1. PolyDoms: Web Resource for Initial Functional Characterization of Protein Polymorphisms Jing Chen1, Sivakumar Gowrisankar1, Anil G Jegga1, Robert Livingston2, Andrew von Niederhausern3, Dana C. Crawford2, Christopher S. Carlson2, Mark J. Rieder2, Robert B. Weiss3, Deborah A. Nickerson2 and Bruce J Aronow1 1Biomedical Informatics, Cincinnati Childrens Hospital Medical Center and University of Cincinnati. 2Department of Genome Sciences, University of Washington, 3Department of Human Genetics, University of Utah Of the 90 NIEHS DNA repair candidate proteins with a known functional domain, 6 of them had a BRCT domain. Abstract NIEHS Candidate Genes: Prioritized List As discoveries of genetic polymorphisms in the human population expand, so does the opportunity and challenge of correlating these with disease-risk. Thus, there is a critical need to efficiently organize large-scale polymorphism analyses and to prioritize their further testing through experimental and epidemiologic studies. To approach this, we have developed a web accessible server PolyDoms (http://polydoms.cchmc.org/) as a resource that can serve as an initial filter for the identification of potentially high impact nsSNPs (nonsynonymous Single Nucleotide Polymorphisms). In particular we have concentrated on potentially deleterious polymorphisms that might affect gene expression, protein function and pathways that could lead to increased environmental agent sensitivity and risk of diseases associated with loss of genomic integrity. We have now mapped all EGP (Environmental Genome Project) cSNPs (coding SNPs) onto the corresponding conserved and known functional protein domains from NCBI’s CDD (Conserved Domain Database) and the 3D structures from PDB (Protein Data Bank). PolyDoms provides an interactive graphical visualization web interface, is easy to update with new polymorphisms and automatically retrieves relevant literature references. Using an automatic link to the Polyview server (http://polyview.cchmc.org), polymorphisms are also mapped to extremely high quality predictions of protein secondary structures and relative solvent accessibilities with polymorphic residues highlighted. This is also available for each of the DNA repair and cell cycle control group of proteins. In addition, the results from nsSNP effect-prediction servers like PolyPhen (Polymorphism Phenotyping) and SIFT (Sorting Intolerant from Tolerant) are also performed automatically and made available for each of the nsSNPs. We are now also extending the horizon of polymorphisms studied to include insertions and deletions of putative regulatory regions—likely to be of much higher regulatory impact than SNPs—of environmentally responsive genes. Significant correlation reported between FGFR4 SNP (Arg 388 Gly) and prognosis in patients with soft tissue sarcoma. This SNP might be used to improve the prediction of clinical prognosis and lead to new treatment strategies in patients with soft tissue sarcomas (Morimoto et al., 2003). CYP2A6*1 (wild type) is responsible for the 7-hydroxylation of coumarin. The point mutation (T to A) in codon 160 leads to a single amino acid substitution (Leu to His) and the resulting protein, CYP2A*2 is unable to 7-hydroxylate coumarin (Cok et al., 2001). Of the 127 NIEHS Cell Cycle candidate proteins with a known functional domain, 9 of them had a transcription factor E2F/dimerization domain. EPHX1 (microsomal epoxide hydrolase) codon 113 Tyr/Tyr variant is associated with oropharyngeal carcinogenesis (Amador et al., 2002). PolyView (http://polyview.cchmc.org) PolyDoms: http://polydoms.cchmc.org Schematic representation of secondary structures, relative solvent accessibilities, highlighting polymorphic residues. The RSA reflects the degree of the residue’s exposure to the surrounding solvent in the protein structure. The relative probability of disease-causing mutations is highest in the protein interior. NIEHS-EGP sequenced genes that have at least one nsSNP occurring in a protein functional domain and predicted as probably damaging or deleterious by PolyPhen and SIFT algorithms respectively. There are no reports of any disease implications of Ala1170Pro nsSNP of ERBB2A though it’s predicted as deleterious and is occurring in a functional protein domain. However, there are conflicting reports about another nsSNP Ile655Val. It has been associated with an increased risk of breast cancer, particularly among younger women. However, this SNP has variable frequency in different ethnic groups (Ameyaw et al., 2002). SIFT and PolyPhen predicted it as benign or tolerated. A RefSeq protein database search for additional protein targets that have E2F/dimerization domain resulted in 4 other proteins (3 hypothetical and an E2F7). E2F7 has a nsSNP (Ala324Asp) occurring in the E2F domain of the protein. Twenty-three of the NIEHS candidate genes have at least one nsSNP occurring in a known protein functional domain and predicted as deleterious or damaging by SIFT and PolyPhen servers respectively. References • PolyDoms: http://polydoms.cchmc.org • PolyView: http://polyview.cchmc.org • SIFT: http://blocks.fhcrc.org/sift/SIFT.html • PolyPhen: http://www.bork.embl-heidelberg.de/PolyPhen/ • GeneSNPs: http://www.genome.utah.edu/genesnps/ • NCBI RefSeq: ftp://ftp.ncbi.nih.gov/refseq/ • NCBI-CDD: http://www.ncbi.nlm.nih.gov/Structure/cdd/cdd.shtml • Mutations at Arg residues account for almost 15% of disease mutations. • A random mutation at a Trp or Cys residue has highest probability of causing disease. • Mutations at Gly which is frequently present at the turns of alpha-helices, might have a negative impact on protein structural stability (Vitkup et al., 2003). 48 of 135 cell cycle proteins had at least one nsSNP affecting an Arg residue; 11-Trp; 15-Cys; and 30 proteins with at least one nsSNP at a glycine residue. Support NIEHS U01 ES11038 Mouse Centers Genomics Consortium