MOTIVATION: Subspecies identification is one of the most critical issues in microbiome studies, as it is directly related to their functions in response to the environmental stress and their feedbacks. However, identification of subspecies remains a challenge largely due to the small variation between different strains within the same species. Accurate identification of subspecies primarily relies on variant identification and categorization through microbiome data. However, current SNP calling and subspecies identification for microbiome data remain underdeveloped. RESULTS: In this work, we have proposed Strain-GeMS for subspecies identification from microbiome data, based on solid statistical model for SNP calling, as well as optimized procedure for subspecies identification. Results on simulated, ab initio and in vivo datasets have shown that Strain-GeMS could always generate more accurate results compared with other subspecies identification methods. AVAILABILITY AND IMPLEMENTATION: Strain-GeMS is available at: https://github.com/HUST-NingKang-Lab/straingems. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Recent advances in next-generation sequencing technology have yielded increasing cost-effectiveness and higher throughput produced per run, in turn, greatly influencing the analysis of DNA sequences. Among the various sequencing technologies, Illumina is by far the most widely used platform. However, the Illumina sequencing platform suffers from several imperfections that can be attributed to the chemical processes inherent to the sequencing-by-synthesis technology. With the enormous amounts of reads produced, statistical methodologies and computationally efficient algorithms are required to improve the accuracy and speed of base-calling. Over the past few years, several papers have proposed methods to model the various imperfections, giving rise to accurate and/or efficient base-calling algorithms. In this article, we provide a comprehensive comparison of the performance of recently developed base-callers and we present a general statistical model that unifies a large majority of these base-callers.

MOTIVATION: Single nucleotide variant (SNV) detection procedures are being utilized as never before to analyze the recent abundance of high-throughput DNA sequencing data, both on single and multiple sample datasets. Building on previously published work with the single sample SNV caller genotype model selection (GeMS), a multiple sample version of GeMS (MultiGeMS) is introduced. Unlike other popular multiple sample SNV callers, the MultiGeMS statistical model accounts for enzymatic substitution sequencing errors. It also addresses the multiple testing problem endemic to multiple sample SNV calling and utilizes high performance computing (HPC) techniques. RESULTS: A simulation study demonstrates that MultiGeMS ranks highest in precision among a selection of popular multiple sample SNV callers, while showing exceptional recall in calling common SNVs. Further, both simulation studies and real data analyses indicate that MultiGeMS is robust to low-quality data. We also demonstrate that accounting for enzymatic substitution sequencing errors not only improves SNV call precision at low mapping quality regions, but also improves recall at reference allele-dominated sites with high mapping quality. AVAILABILITY AND IMPLEMENTATION: The MultiGeMS package can be downloaded from https://github.com/cui-lab/multigems CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: A review of the available single nucleotide polymorphism (SNP) calling procedures for Illumina high-throughput sequencing (HTS) platform data reveals that most rely mainly on base-calling and mapping qualities as sources of error when calling SNPs. Thus, errors not involved in base-calling or alignment, such as those in genomic sample preparation, are not accounted for. RESULTS: A novel method of consensus and SNP calling, Genotype Model Selection (GeMS), is given which accounts for the errors that occur during the preparation of the genomic sample. Simulations and real data analyses indicate that GeMS has the best performance balance of sensitivity and positive predictive value among the tested SNP callers. AVAILABILITY: The GeMS package can be downloaded from https://sites.google.com/a/bioinformatics.ucr.edu/xinping-cui/home/software or http://computationalbioenergy.org/software.html. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The Affymetrix GeneChip microarray is currently providing a high-density and economical platform for discovery of genetic polymorphisms. Microarray data for single feature polymorphism (SFP) detection in recombinant inbred lines (RILs) can capitalize on the high level of replication available for each locus in the RIL population. It was suggested that the binding affinities from all of the RILs would form a multimodal distribution for a SFP. This motivated us to estimate the binding affinities from the robust multi-array analysis (RMA) method and formulate the SFP detection problem as a hypothesis testing problem, i.e. testing whether the underlying distribution of the estimated binding affinity (EBA) values of a probe is unimodal or multimodal. RESULTS: We developed a bootstrap-based hypothesis testing procedure using the 'dip' statistic. Our simulation studies show that the proposed procedure can reach satisfactory detection power with false discovery rate controlled at a desired level and is robust to the unimodal distribution assumption, which facilitates wide application of the proposed procedure. Our analysis of the real data identified more than four times the SFPs compared to the previous studies, covering 96% of their findings. The constructed genetic map using the SFP markers predicted from our procedure shows over 99% concordance of the genetic orders of these markers with their known physical locations on the genome sequence. AVAILABILITY: The R package 'dipSFP' can be downloaded from http://sites.google.com/a/bioinformatics.ucr.edu/xinping-cui/home/software.

MOTIVATION: Oligonucleotide arrays such as Affymetrix GeneChips use multiple probes, or a probe set, to measure the abundance of mRNA of every gene of interest. Some analysis methods attempt to summarize the multiple observations into one single score before conducting further analysis such as detecting differentially expressed genes (DEG), clustering and classification. However, there is a risk of losing a significant amount of information and consequently reaching inaccurate or even incorrect conclusions during this data reduction. RESULTS: We developed a novel statistical method called robustified multivariate analysis of variance (MANOVA) based on the traditional MANOVA model and permutation test to detect DEG for both one-way and two-way cases. It can be extended to detect some special patterns of gene expression through profile analysis across k (>or=2) populations. The method utilizes probe-level data and requires no assumptions about the distribution of the dataset. We also propose a method of estimating the null distribution using quantile normalization in contrast to the 'pooling' method (Section 3.1). Monte Carlo simulation and real data analysis are conducted to demonstrate the performance of the proposed method comparing with the 'pooling' method and the usual Analysis of Variance (ANOVA) test based on the summarized scores. It is found that the new method successfully detects DEG under desired false discovery rate and is more powerful than the competing method especially when the number of groups is small. AVAILABILITY: The package of robustified MANOVA can be downloaded from http://faculty.ucr.edu/~xpcui/software

MOTIVATION: Genomic DNA was hybridized to oligonucleotide microarrays to identify single-feature polymorphisms (SFP) for Arabidopsis, which has a genome size of approximately 130 Mb. However, that method does not work well for organisms such as barley, with a much larger 5200 Mb genome. In the present study, we demonstrate SFP detection using a small number of replicate datasets and complex RNA as a surrogate for barley DNA. To identify single probes defining SFPs in the data, we developed a method using robustified projection pursuit (RPP). This method first evaluates, for each probe set, the overall differentiation of signal intensities between two genotypes and then measures the contribution of the individual probes within the probe set to the overall differentiation. RESULTS: RNA from whole seedlings with and without dehydration stress provided 'present' calls for approximately 75% of probe sets. Using triplicated data, among the 5% of 'present' probe sets identified as most likely to contain at least one SFP probe, at least 80% are correctly predicted. This was determined by direct sequencing of PCR amplicons derived from barley genomic DNA. Using a 5 percentile cutoff, we defined 2007 SFP probes contained in 1684 probe sets by combining three parental genotype comparisons: Steptoe versus Morex, Morex versus Barke and Oregon Wolfe Barley Dominant versus Recessive. AVAILABILITY: The algorithm is available upon request from the corresponding author. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: http://faculty.ucr.edu/~xpcui.