Network-augmented analysis of disease genomics data
Abstract
Recent genomic revolution opened new avenues to understanding human disease. However, it also revealed complex nature of human disease. For example, currently more than several hundred genes are believed to be associated to human cancer. Genome-wide association study (GWAS) suggests hundreds of disease-related genes, but together explaining only 10-20% of total disease inheritance at most. Because of this overwhelming complexity of disease-causing pathway, modern disease genetics needs to be more systematic and predictive. However, the network organization of disease systems also provide big opportunity to investigate the genetic organization of complex diseases through the molecular networks. Our research group has developed co-functional gene networks for many organisms including human (HumanNet) and various network-guided methods to identify novel disease genes and modules. In this talk, I will present our recent work in network-based augmenting and interpreting cancer somatic mutation data (MUFFINN), GWAS data for complex diseases (GWAB), and gene set enrichment analysis for disease transcriptome data (NGSEA).
Splicing bridging transcriptomics and proteomics in breast cancer
Abstract
Since approximately 90% of human genes produce multiple transcript isoforms, resulting from various combinations of exons (i.e., exon selection) via alternative splicing (AS) mechanisms, many key proteins associated with tumor biology including proteins with roles in apoptosis, cell cycle regulation, invasion and metastasis undergo cancer-associated alternative splicing, which miss a specific exon or retain an undesired intron sequence leading to change of protein product. Thus, cancer-specific splice variants, which may give survival advantages to cancer cells, often result in poor outcomes. Furthermore, AS is becoming an interesting target for drug development and a novel therapeutic strategy in diverse complex diseases including cancer. Here, we summarize our studies to provide an insight on how the splicing can bridge genomics, transcriptomics, and proteomics.
SAAVpedia : Identification, functional annotation, retrieval and prioritization of single amino-acid variants from proteomic and genomic data
Abstract
Through high-throughput genome sequencing, numerous disease/drug associated non-synonymous single nucleotide variants (nsSNVs) that alter the amino acid sequence of a protein have been identified. Only a few nsSNVs could be confirmed in protein level by proteogenomic analysis that named single amino acid variants (SAAVs). Even though some studies tried to characterize and pinpoint pathogenic SAAVs, most of them remain uncharacterized. Here, we developed the SAAVpedia to identify, annotate, retrieve and prioritize pathogenic SAAVs from proteomic data. The SAAVpedia provides reference protein sequence database which contained 5,324,509 SAAVs in 42,134 protein isoforms. The annotation database contained ten kinds of comprehensive biological, clinical and pharmacological information. It enables ranged queries to search for condition-specific SAAVs that related to post-translational modifications (PTM) such as glycosylation, acetylation, etc. It can also retrieve the SAAVs that associate with disease, drug, genes, genomic location and reported information on the 1000 Genome, dbSNP and COSMIC. We also applied a new pathogenic prioritization algorithm using 11,891 SAAVs from proteomic data of breast cancer patients. SAAVpedia (http://SAAVpedia.org) will be provided as a python standalone and a web application for convenience of users.
Rm#422, Industry-University Research Bldg., Yonsei University
50 Yonsei-ro, Seodeamoon-gu, Seoul 03722 Korea
Tel : +82-2-393-8328 Fax : +82-2-393-6589 E-mail : admin@khupo.org
COPYRIGHT ⓒ 2018 KHUPO. ALL RIGHTS RESERVED.
