Dataset Information

Identifying hypothetical genetic influences on complex disease phenotypes.

ABSTRACT: Statistical interactions between disease-associated loci of complex genetic diseases suggest that genes from these regions are involved in a common mechanism impacting, or impacted by, the disease. The computational problem we address is to discover relationships among genes from these interacting regions that may explain the observed statistical interaction and the role of these genes in the disease phenotype.We describe a heuristic algorithm for generating hypothetical gene relationships from loci associated with a complex disease phenotype. This approach, called Prioritizing Disease Genes by Analysis of Common Elements (PDG-ACE), mines biomedical keywords from text descriptions of genes and uses them to relate genes close to disease-associated loci. A keyword common to, and significantly over-represented in, a pair of gene descriptions may represent a preliminary hypothesis about the biological relationship between the genes, and suggest the role the genes play in the disease phenotype.Our experimentation shows that the approach finds previously published relationships, while failing to find relationships that don't exist. The results also indicate that the approach is robust to differences in keyword vocabulary. We outline a brief case study in which results from a recently published Type 2 Diabetes association study are used to identify potential hypotheses.

SUBMITTER: Keller BJ

PROVIDER: S-EPMC2646236 | biostudies-literature | 2009 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Identifying hypothetical genetic influences on complex disease phenotypes.

Keller Benjamin J BJ McEachin Richard C RC

BMC bioinformatics 20090205

<h4>Background</h4>Statistical interactions between disease-associated loci of complex genetic diseases suggest that genes from these regions are involved in a common mechanism impacting, or impacted by, the disease. The computational problem we address is to discover relationships among genes from these interacting regions that may explain the observed statistical interaction and the role of these genes in the disease phenotype.<h4>Results</h4>We describe a heuristic algorithm for generating hy ...[more]

PMID: 19208188

Similar Datasets

Project description:BackgroundA critical goal in biology is to relate the phenotype to the genotype, that is, to find the genetic determinants of various traits. However, while simple monofactorial determinants are relatively easy to identify, the underpinnings of complex phenotypes are harder to predict. While traditional approaches rely on genome-wide association studies based on Single Nucleotide Polymorphism data, the ability of machine learning algorithms to find these determinants in whole proteome data is still not well known.ResultsTo better understand the applicability of machine learning in this case, we implemented two such algorithms, adaptive boosting (AB) and repeated random forest (RRF), and developed a chunking layer that facilitates the analysis of whole proteome data. We first assessed the performance of these algorithms and tuned them on an influenza data set, for which the determinants of three complex phenotypes (infectivity, transmissibility, and pathogenicity) are known based on experimental evidence. This allowed us to show that chunking improves runtimes by an order of magnitude. Based on simulations, we showed that chunking also increases sensitivity of the predictions, reaching 100% with as few as 20 sequences in a small proteome as in the influenza case (5k sites), but may require at least 30 sequences to reach 90% on larger alignments (500k sites). While RRF has less specificity than random forest, it was never <50%, and RRF sensitivity was significantly higher at smaller chunk sizes. We then used these algorithms to predict the determinants of three types of drug resistance (to Ciprofloxacin, Ceftazidime, and Gentamicin) in a bacterium, Pseudomonas aeruginosa. While both algorithms performed well in the case of the influenza data, results were more nuanced in the bacterial case, with RRF making more sensible predictions, with smaller errors rates, than AB.ConclusionsAltogether, we demonstrated that ML algorithms can be used to identify genetic determinants in small proteomes (viruses), even when trained on small numbers of individuals. We further showed that our RRF algorithm may deserve more scrutiny, which should be facilitated by the decreasing costs of both sequencing and phenotyping of large cohorts of individuals.

Dataset Information

Identifying hypothetical genetic influences on complex disease phenotypes.

Publications

Identifying hypothetical genetic influences on complex disease phenotypes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets