Selection of Multiple SNPs in Case-Control Association Study Using a Discretized Network Flow Approach

Abstract

Abstract. Recent large scale genome-wide association studies have been considered to hold promise for unraveling the genetic etiology of complex diseases. It becomes possible now to use these data to assess the influence of interactions from multiple SNPs on a disease. In this paper we formulate the multiple SNP selection problem for determining genetic risk profiles of certain diseases by formulating novel 0/1 IP formulations for this problem, and solving them using a new near-optimal and efficient discrete optimization technique called discretized network flow that has recently been developed by us. One of the highlights of our approach to solving the multiple SNP selection problem is recognizing that there could be different genetic profiles of a disease among the patient population, and it is thus desirable to classify/cluster patients with similar genetic profiles of the disease while simultaneously selecting the right genetic marker sets of the disease for each cluster. This approach coupled with the DNF technique has yielded results for several diseases with some of the highest sensitivities seen so far and specificities that are higher or comparable to state-of-the art techniques, at a fraction of the runtime of these techniques.

Discovery of protein acetylation patterns by deconvolution of peptide isomer mass spectra

“…In Iso-PeptidAce, fragment ion patterns are reduced to a set of maximum network flow problems, for which a number of efficient algorithms are known 15 . This approach has previously been used in a wide range of complex problems, such as predicting molecular pathways in complex diseases or selecting single nucleotide polymorphisms and their associated alleles in patient and control groups 16 17 18 . In our study, we implemented the network flow approach to deconvolute composite fragment spectra of acetylated isomers.…”

USVM: Selection of SNPs in Diseases Association Study Using UMDA and SVM

“…From the table we can see that the prediction accuracy is high (greater than 90%) under LOO. Finally, we compare our algorithm with several current methods including DNF [16], CGSP [19], RCGSP [19], CSP [18] and ORF [17]. The results of DNF were achieved by the 5-fold cross validation procedure, while the CGSP、RCGSP and CSP results were from LOO cross validation procedure.…”

“…Dutt formulated the multi-SNP selection problem as 0/1 integer programming problem, then solved them using a near-optimal technique called discretized network flow [16]. Mao and Lee used random forest to choose most disease-associated SNPs, and then selected SNPs used to build the optimal random forest to predict the disease status [17].…”