UNLABELLED: We present a new R package for training gapped-kmer SVM classifiers for DNA and protein sequences. We describe an improved algorithm for kernel matrix calculation that speeds run time by about 2 to 5-fold over our original gkmSVM algorithm. This package supports several sequence kernels, including: gkmSVM, kmer-SVM, mismatch kernel and wildcard kernel. AVAILABILITY AND IMPLEMENTATION: gkmSVM package is freely available through the Comprehensive R Archive Network (CRAN), for Linux, Mac OS and Windows platforms. The C ++ implementation is available at www.beerlab.org/gkmsvm CONTACT: [email protected] or [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: RNAs play fundamental roles in cellular processes. The function of an RNA is highly dependent on its 3D conformation, which is referred to as the RNA tertiary structure. Because the prediction or experimental determination of these structures is difficult, so many works focus on the problems associated with the RNA secondary structure. Here, we consider the RNA inverse folding problem, in which an RNA secondary structure is given as a target structure and the goal is to design an RNA sequence that folds into the target structure. In this article, we introduce a new evolutionary algorithm for the RNA inverse folding problem. Our algorithm, entitled Evolutionary RNA Design, generates a sequence whose minimum free energy structure is the same as the target structure. RESULTS: We compare our algorithm with INFO-RNA, MODENA, RNAiFold and NUPACK approaches for some biological test sets. The results presented in this article indicate that for longer structures, our algorithm performs better than the other mentioned algorithms in terms of the energy range, accuracy, speedup and nucleotide distribution. Particularly, the generated RNA sequences in our method are much more reliable and similar to the natural RNA sequences.

Oligomers of length k, or k-mers, are convenient and widely used features for modeling the properties and functions of DNA and protein sequences. However, k-mers suffer from the inherent limitation that if the parameter k is increased to resolve longer features, the probability of observing any specific k-mer becomes very small, and k-mer counts approach a binary variable, with most k-mers absent and a few present once. Thus, any statistical learning approach using k-mers as features becomes susceptible to noisy training set k-mer frequencies once k becomes large. To address this problem, we introduce alternative feature sets using gapped k-mers, a new classifier, gkm-SVM, and a general method for robust estimation of k-mer frequencies. To make the method applicable to large-scale genome wide applications, we develop an efficient tree data structure for computing the kernel matrix. We show that compared to our original kmer-SVM and alternative approaches, our gkm-SVM predicts functional genomic regulatory elements and tissue specific enhancers with significantly improved accuracy, increasing the precision by up to a factor of two. We then show that gkm-SVM consistently outperforms kmer-SVM on human ENCODE ChIP-seq datasets, and further demonstrate the general utility of our method using a Naïve-Bayes classifier. Although developed for regulatory sequence analysis, these methods can be applied to any sequence classification problem.

Reconstructing haplotype in MEC (minimum error correction) model is an important clustering problem which focuses on inferring two haplotypes from SNP fragments (single nucleotide polymorphism) containing gaps and errors. Mutated form of human genome is responsible for genetic diseases which mostly occur in SNP sites. In this paper, a fuzzy clustering approach is performed for haplotype reconstruction or haplotype assembly from a given sample single nucleotide polymorphism (SNP). In the best previous approach based on reconstruction rate (Wang, 2007), all SNP-fragments are considered with equal values. In our proposed method the value of the fragments are based on the degree of membership between SNP-fragments and centers of clusters. Finally, these two approaches are executed on four standard datasets (ACE, Daly, SIM0 and SIM50) and the results show the efficiency of our proposed approach.