Search

Ongoing increases in the size of human genotype and phenotype collections offer the promise of improved understanding of the genetics of complex diseases. In addition to the biological insights that can be gained from the nature of variants that contribute to the genetic component of complex trait variability, these data have brought forward the prospect of predicting complex traits and the risk of complex genetic diseases from genotype data. Optimal realization of these objectives requires ongoing methodological developments, designed to identify true trait-associated variants and accurately predict phenotype from genotype. In addition to optimal accuracy, these methods must also be computationally efficient, in order to remain tractable in the context of high variant densities and very large sample sizes. Here we show that the power of linear mixed models that are in widespread use for GWAS can be increased significantly by modeling off target genetic effects using polygenic scores derived from SNPs that are not on the same chromosome as the target SNP. Applying this method to simulated and real data we found that it can result in a substantial increase in the number of variants passing genome-wide significance thresholds. This increase in power to detect trait-associated variants also translates into an increase in the accuracy with which the resulting polygenic score predicts the phenotype from genotype data.

Diploidy has profound implications for population genetics and susceptibility to genetic diseases. Although two copies are present for most genes in the human genome, they are not necessarily both active or active at the same level in a given individual. Genomic imprinting, resulting in exclusive or biased expression in favor of the allele of paternal or maternal origin, is now believed to affect hundreds of human genes. A far greater number of genes display unequal expression of gene copies due to cis-acting genetic variants that perturb gene expression. The availability of data generated by RNA sequencing applied to large numbers of individuals and tissue types has generated unprecedented opportunities to assess the contribution of genetic variation to allelic imbalance in gene expression. Here we review the insights gained through the analysis of these data about the extent of the genetic contribution to allelic expression imbalance, the tools and statistical models for gene expression imbalance, and what the results obtained reveal about the contribution of genetic variants that alter gene expression to complex human diseases and phenotypes. C.S. and S.C. are funded by Science Foundation Ireland grant number 16/IA/4612. peer-reviewed

Background Tumour mutation burden (TMB), defined as the number of somatic mutations per megabase within the sequenced region in the tumour sample, has been used as a biomarker for predicting response to immune therapy. Several studies have been conducted to assess the utility of TMB for various cancer types; however, methods to measure TMB have not been adequately evaluated. In this study, we identified two sources of bias in current methods to calculate TMB. Methods We used simulated data to quantify the two sources of bias and their effect on TMB calculation, we down-sampled sequencing reads from exome sequencing datasets from TCGA to evaluate the consistency in TMB estimation across different sequencing depths. We analyzed data from ten cancer cohorts to investigate the relationship between inferred TMB and sequencing depth. Results We found that TMB, estimated by counting the number of somatic mutations above a threshold frequency (typically 0.05), is not robust to sequencing depth. Furthermore, we show that, because only mutations with an observed frequency greater than the threshold are considered, the observed mutant allele frequency provides a biased estimate of the true frequency. This can result in substantial over-estimation of the TMB, when the cancer sample includes a large number of somatic mutations at low frequencies, and exacerbates the lack of robustness of TMB to variation in sequencing depth and tumour purity. Conclusion Our results demonstrate that care needs to be taken in the estimation of TMB to ensure that results are unbiased and consistent across studies and we suggest that accurate and robust estimation of TMB could be achieved using statistical models that estimate the full mutant allele frequency spectrum.

Phylogenetic models of the evolution of protein-coding sequences can provide insights into the selection pressures that have shaped them. In the application of these models synonymous nucleotide substitutions, which do not alter the encoded amino acid, are often assumed to have limited functional consequences and used as a proxy for the neutral rate of evolution. The ratio of nonsynonymous to synonymous substitution rates is then used to categorize the selective regime that applies to the protein (e.g. purifying selection, neutral evolution, diversifying selection). Here, we extend these models to explore the extent of purifying selection acting on substitutions between synonymous stop codons. Using a large collection of coding sequence alignments we estimate that a high proportion (approximately 57%) of mammalian genes are affected by selection acting on stop codon preference. This proportion varies substantially by codon, with UGA stop codons far more likely to be conserved. Genes with evidence of selection acting on synonymous stop codons have distinctive characteristics, compared to unconserved genes with the same stop codon, including enrichment for specific functional properties. Notably, genes with conserved stop codons have longer 3′ UTRs and are associated with shorter mRNA half-life than other genes. The coding regions of these genes are also much more likely to be under strong purifying selection pressure. Our results suggest that the preference for UGA stop codons in many multicellular eukaryotes is selective rather than mutational in origin. Author summary Three codons can specify the termination of translation of protein-coding genes, but the efficiency of translation termination differs between them. Readthrough translation, in which the ribosome fails to terminate at the stop codon, can result in aberrant proteins but can also have a regulatory role. By extending statistical models used to study the evolution of sense codons, we performed a large-scale analysis of selection acting on mutations between the three stop codons in mammals. We estimate that more than half of mammalian protein-coding genes are affected by selection that reduces the rate of substitution between synonymous stop codons, compared to what we would expect if the stop codons were functionally equivalent. Our results suggest that the choice of stop codon is affected by evolutionary selection and impacts the function or regulation of a substantial proportion of mammalian genes.

Summary DNA replication initiates from multiple origins, and selective CDC7 kinase inhibitors (CDC7is) restrain cell proliferation by limiting origin firing. We have performed a CRISPR-Cas9 genome-wide screen to identify genes that, when lost, promote the proliferation of cells treated with sub-efficacious doses of a CDC7i. We have found that the loss of function of ETAA1, an ATR activator, and RIF1 reduce the sensitivity to CDC7is by allowing DNA synthesis to occur more efficiently, notably during late S phase. We show that partial CDC7 inhibition induces ATR mainly through ETAA1, and that if ATR is subsequently inhibited, origin firing is unleashed in a CDK- and CDC7-dependent manner. Cells are then driven into a premature and highly defective mitosis, a phenotype that can be recapitulated by ETAA1 and TOPBP1 co-depletion. This work defines how ATR mediates the effects of CDC7 inhibition, establishing the framework to understand how the origin firing checkpoint functions.

Expression quantitative trait loci (eQTLs) identified using tumor gene expression data could affect gene expression in cancer cells, tumor-associated normal cells, or both. Here, we have demonstrated a method to identify eQTLs affecting expression in cancer cells by modeling the statistical interaction between genotype and tumor purity. Only one third of breast cancer risk variants, identified as eQTLs from a conventional analysis, could be confidently attributed to cancer cells. The remaining variants could affect cells of the tumor microenvironment, such as immune cells and fibroblasts. Deconvolution of tumor eQTLs will help determine how inherited polymorphisms influence cancer risk, development, and treatment response. Electronic supplementary material The online version of this article (10.1186/s13059-018-1507-0) contains supplementary material, which is available to authorized users.

Do human societies from around the world exhibit similarities in the way that they are structured, and show commonalities in the ways that they have evolved? These are long-standing questions that have proven difficult to answer. To test between competing hypotheses, we constructed a massive repository of historical and archaeological information known as "Seshat: Global History Databank." We systematically coded data on 414 societies from 30 regions around the world spanning the last 10,000 years. We were able to capture information on 51 variables reflecting nine characteristics of human societies, such as social scale, economy, features of governance, and information systems. Our analyses revealed that these different characteristics show strong relationships with each other and that a single principal component captures around three-quarters of the observed variation. Furthermore, we found that different characteristics of social complexity are highly predictable across different world regions. These results suggest that key aspects of social organization are functionally related and do indeed coevolve in predictable ways. Our findings highlight the power of the sciences and humanities working together to rigorously test hypotheses about general rules that may have shaped human history.

Cell subtype proportion variability between samples contributes significantly to the variation of functional genomic properties such as gene expression or DNA methylation. Although the impact of the variation of cell subtype composition on measured genomic quantities is recognized, and some innovative tools have been developed for the analysis of heterogeneous samples, most functional genomics studies using samples with mixed cell types still ignore the influence of cell subtype proportion variation, or just deal with it as a nuisance variable to be eliminated. Here we demonstrate how harvesting information about cell subtype proportions from functional genomics data can provide insights into cellular changes associated with phenotypes. We focused on two types of mixed cell populations, human blood and mouse kidney. Cell type prediction is well developed in the former, but not currently in the latter. Estimating the cellular repertoire is easier when a reference dataset from purified samples of all cell types in the tissue is available, as is the case for blood. However, reference datasets are not available for most other tissues, such as the kidney. In this study, we showed that the proportion of alterations attributable to changes in the cellular composition varies strikingly in the two disorders (asthma and systemic lupus erythematosus), suggesting that the contribution of cell subtype proportion changes to functional genomic properties can be disease-specific. We also showed that a reference dataset from a single-cell RNA-seq study successfully estimated the cell subtype proportions in mouse kidney and allowed us to distinguish altered cell subtype differences between two different knock-out mouse models, both of which had reported a reduced number of glomeruli compared to their wild-type counterparts. These findings demonstrate that testing for changes in cell subtype proportions between conditions can yield important insights in functional genomics studies.

Aberrant DNA methylation patterns have been reported in inflamed tissues and may play a role in disease. We studied DNA methylation and gene expression profiles of purified intestinal epithelial cells from ulcerative colitis patients, comparing inflamed and non-inflamed areas of the colon. We identified 577 differentially methylated sites (false discovery rate

The rate of germline mutation varies widely between species but little is known about the extent of variation in the germline mutation rate between individuals of the same species. Here we demonstrate that an allele that increases the rate of germline mutation can result in a distinctive signature in the genomic region linked to the affected locus, characterized by a number of haplotypes with a locally high proportion of derived alleles, against a background of haplotypes carrying a typical proportion of derived alleles. We searched for this signature in human haplotype data from phase 3 of the 1000 Genomes Project and report a number of candidate mutator loci, several of which are located close to or within genes involved in DNA repair or the DNA damage response. To investigate whether mutator alleles remained active at any of these loci, we used de novo mutation counts from human parent-offspring trios in the 1000 Genomes and Genome of the Netherlands cohorts, looking for an elevated number of de novo mutations in the offspring of parents carrying a candidate mutator haplotype at each of these loci. We found some support for two of the candidate loci, including one locus just upstream of the BRSK2 gene, which is expressed in the testis and has been reported to be involved in the response to DNA damage., Author Summary Each time a genome is replicated there is the possibility of error resulting in the incorporation of an incorrect base or bases in the genome sequence. When these errors occur in cells that lead to the production of gametes they can be incorporated into the germline. Such germline mutations are the basis of evolutionary change; however, to date there has been little attempt to quantify the extent of genetic variation in human populations in the rate at which they occur. This is particularly important because new spontaneous mutations are thought to make an important contribution to many human diseases. Here we present a new way to identify genetic loci that may be associated with an elevated rate of germline mutation and report the application of this method to data from a large number of human genomes, generated by the 1000 Genomes Project. Several of the candidate loci we report are in or near genes involved in DNA repair.

We share the view of Dr. Watson and colleagues on the importance of germline databases (GLDB) of immune receptor genes and thank them for their interest in our work ([1][1]). Whereas we fully agree with them on many of the caveats that apply to alleles inferred from genome sequence data (discussed

Nonnegative Matrix Factorization (NMF) has proved to be an effective method for unsupervised clustering analysis of gene expression data. By the nonnegativity constraint, NMF provides a decomposition of the data matrix into two matrices that have been used for clustering analysis. However, the decomposition is not unique. This allows different clustering results to be obtained, resulting in different interpretations of the decomposition. To alleviate this problem, some existing methods directly enforce uniqueness to some extent by adding regularization terms in the NMF objective function. Alternatively, various normalization methods have been applied to the factor matrices; however, the effects of the choice of normalization have not been carefully investigated. Here we investigate the performance of NMF for the task of cancer class discovery, under a wide range of normalization choices. After extensive evaluations, we observe that the maximum norm showed the best performance, although the maximum norm has not previously been used for NMF. Matlab codes are freely available from: http://maths.nuigalway.ie/~haixuanyang/pNMF/pNMF.htm.

Region and cell-type specific differences in the molecular make up of colon epithelial cells have been reported. Those differences may underlie the region-specific characteristics of common colon epithelial diseases such as colorectal cancer and inflammatory bowel disease. DNA methylation is a cell-type specific epigenetic mark, essential for transcriptional regulation, silencing of repetitive DNA and genomic imprinting. Little is known about any region-specific variations in methylation patterns in human colon epithelial cells. Using purified epithelial cells and whole biopsies ( n = 19) from human subjects, we generated epigenome-wide DNA methylation data (using the HELP-tagging assay), comparing the methylation signatures of the proximal and distal colon. We identified a total of 125 differentially methylated sites (DMS) mapping to transcription start sites of protein-coding genes, most notably several members of the homeobox ( HOX) family of genes. Patterns of differential methylation were validated with MassArray EpiTYPER. We also examined DNA methylation in whole biopsies, applying a computational technique to deconvolve variation in methylation within cell types and variation in cell-type composition across biopsies. Including inferred epithelial proportions as a covariate in differential methylation analysis applied to the whole biopsies resulted in greater overlap with the results obtained from purified epithelial cells compared with when the covariate was not included. Results obtained from both approaches highlight region-specific methylation patterns of HOX genes in colonic epithelium. Regional variation in methylation patterns has implications for the study of diseases that exhibit regional expression patterns in the human colon, such as inflammatory bowel disease and colorectal cancer.

Admixed populations present unique opportunities to discover the genetic factors underlying many multifactorial diseases. The geographical position and complex history of South Africa has led to the establishment of the unique admixed population known as the South African Coloured. Not much is known about the genetic make-up of this population, and the historical record is patchy. We genotyped 959 individuals from the Western Cape area, self-identified as belonging to this population, using the Affymetrix 500k genotyping platform. This resulted in nearly 75,000 autosomal SNPs that could be compared with populations represented in the International HapMap Project and the Human Genome Diversity Project. Analysis by means of both the admixture and linkage models in STRUCTURE revealed that the major ancestral components of this population are predominantly Khoesan (32-43%), Bantu-speaking Africans (20-36%), European (21-28%) and a smaller Asian contribution (9-11%), depending on the model used. This is consistent with historical data. While of great historical and genealogical interest, this information is also essential for future admixture mapping of disease genes in this population.

Existing methods for the prediction of immunologically active T-cell epitopes are based on the amino acid sequence or structure of pathogen proteins. Additional information regarding the locations of epitopes may be acquired by considering the evolution of viruses in hosts with different immune backgrounds. In particular, immune-dependent evolutionary patterns at sites within or near T-cell epitopes can be used to enhance epitope identification. We have developed a mutation-selection model of T-cell epitope evolution that allows the human leukocyte antigen (HLA) genotype of the host to influence the evolutionary process. This is one of the first examples of the incorporation of environmental parameters into a phylogenetic model and has many other potential applications where the selection pressures exerted on an organism can be related directly to environmental factors. We combine this novel evolutionary model with a hidden Markov model to identify contiguous amino acid positions that appear to evolve under immune pressure in the presence of specific host immune alleles and that therefore represent potential epitopes. This phylogenetic hidden Markov model provides a rigorous probabilistic framework that can be combined with sequence or structural information to improve epitope prediction. As a demonstration, we apply the model to a data set of HIV-1 protein-coding sequences and host HLA genotypes.

It is unresolved whether recently transmitted human immunodeficiency viruses (HIV) have genetic features that specifically favour their transmissibility. To identify potential “transmission signatures”, we compared 20 full-length HIV-1 subtype C genomes from primary infections, with 66 sampled from ethnically and geographically matched individuals with chronic infections. Controlling for recombination and phylogenetic relatedness, we identified 39 sites at which amino acid frequency spectra differed significantly between groups. These sites were predominantly located within Env, Pol and Gag (14/39, 9/39 and 6/39 respectively) and were significantly clustered (33/39) within known immunoreactive peptides. Within 6 months of infection, we detected reversion-to-consensus mutations at 14 sites and potential CTL escape mutations at seven. Here we provide evidence that frequent reversion mutations probably allows the virus to recover replicative fitness which, together with immune escape driven by the HLA alleles of the new hosts, differentiate sequences from chronic infections from those sampled shortly after transmission.

It is unknown whether patterns of human immunodeficiency virus (HIV)-specific T-cell responses during acute infection may influence the viral set point and the course of disease. We wished to establish whether the magnitude and breadth of HIV type 1 (HIV-1)-specific T-cell responses at 3 months postinfection were correlated with the viral-load set point at 12 months and hypothesized that the magnitude and breadth of HIV-specific T-cell responses during primary infection would predict the set point. Gamma interferon (IFN-γ) enzyme-linked immunospot (ELISPOT) assay responses across the complete proteome were measured in 47 subtype C HIV-1-infected participants at a median of 12 weeks postinfection. When corrected for amino acid length and individuals responding to each region, the order of recognition was as follows: Nef > Gag > Pol > Rev > Vpr > Env > Vpu > Vif > Tat. Nef responses were significantly ( P < 0.05) dominant, targeted six epitopic regions, and were unrelated to the course of viremia. There was no significant difference in the magnitude and breadth of responses for each protein region with disease progression, although there was a trend of increased breadth (mean, four to seven pools) in rapid progressors. Correlation of the magnitude and breadth of IFN-γ responses with the viral set point at 12 months revealed almost zero association for each protein region. Taken together, these data demonstrate that the magnitude and breadth of IFN-γ ELISPOT assay responses at 3 months postinfection are unrelated to the course of disease in the first year of infection and are not associated with, and have low predictive power for, the viral set point at 12 months.

The adaptive immune system includes populations of B and T cells capable of binding foreign epitopes via antigen specific receptors, called immunoglobulin (IG) for B cells and the T cell receptor (TCR) for T cells. In order to provide protection from a wide range of pathogens, these cells display highly diverse repertoires of IGs and TCRs. This is achieved through combinatorial rearrangement of multiple gene segments in addition, for B cells, to somatic hypermutation. Deep sequencing technologies have revolutionized analysis of the diversity of these repertoires; however, accurate TCR/IG diversity profiling requires specialist bioinformatics tools. Here we present LymAnalzyer, a software package that significantly improves the completeness and accuracy of TCR/ IG profiling from deep sequence data and includes procedures to identify novel alleles of gene segments. On real and simulated data sets LymAnalyzer produces highly accurate and complete results. Although, to date we have applied it to TCR/ IG data from human and mouse, it can be applied to data from any species for which an appropriate database of reference genes is available. Implemented in Java, it includes both a command line version and a graphical user interface and is freely available at https://sourceforge.net/projects/lymanalyzer/.

Background Quantifying gene expression by RNA-Seq has several advantages over microarrays, including greater dynamic range and gene expression estimates on an absolute, rather than a relative scale. Nevertheless, microarrays remain in widespread use, demonstrated by the ever-growing numbers of samples deposited in public repositories. Results We propose a novel approach to microarray analysis that attains many of the advantages of RNA-Seq. This method, called Machine Learning of Transcript Expression (MaLTE), leverages samples for which both microarray and RNA-Seq data are available, using a Random Forest to learn the relationship between the fluorescence intensity of sets of microarray probes and RNA-Seq transcript expression estimates. We trained MaLTE on data from the Genotype-Tissue Expression (GTEx) project, consisting of Affymetrix gene arrays and RNA-Seq from over 700 samples across a broad range of human tissues. Conclusion This approach can be used to accurately estimate absolute expression levels from microarray data, at both gene and transcript level, which has not previously been possible. This methodology will facilitate re-analysis of archived microarray data and broaden the utility of the vast quantities of data still being generated. Electronic supplementary material The online version of this article (doi:10.1186/s12859-015-0712-z) contains supplementary material, which is available to authorized users.

We report a study of genome-wide, dense SNP (∼900K) and copy number polymorphism data of indigenous southern Africans. We demonstrate the genetic contribution to southern and eastern African populations, which involved admixture between indigenous San, Niger-Congo-speaking and populations of Eurasian ancestry. This finding illustrates the need to account for stratification in genome-wide association studies, and that admixture mapping would likely be a successful approach in these populations. We developed a strategy to detect the signature of selection prior to and following putative admixture events. Several genomic regions show an unusual excess of Niger-Kordofanian, and unusual deficiency of both San and Eurasian ancestry, which were considered the footprints of selection after population admixture. Several SNPs with strong allele frequency differences were observed predominantly between the admixed indigenous southern African populations, and their ancestral Eurasian populations. Interestingly, many candidate genes, which were identified within the genomic regions showing signals for selection, were associated with southern African-specific high-risk, mostly communicable diseases, such as malaria, influenza, tuberculosis, and human immunodeficiency virus/AIDs. This observation suggests a potentially important role that these genes might have played in adapting to the environment. Additionally, our analyses of haplotype structure, linkage disequilibrium, recombination, copy number variation and genome-wide admixture highlight, and support the unique position of San relative to both African and non-African populations. This study contributes to a better understanding of population ancestry and selection in south-eastern African populations; and the data and results obtained will support research into the genetic contributions to infectious as well as non-communicable diseases in the region., Author Summary Genome-wide analysis of human populations is useful in shedding light on the evolutionary history of the human genome, with a wide range of applications from reconstructing past associations between different population histories to disease mapping. In this manuscript we report on the application of genome-wide data to southern African populations and the identification of genome-wide signatures of selection pre- and post-admixture. Several signals of selection, before and after admixture, were identified, some of which involved loci associated with human diseases, including malaria, influenza, tuberculosis and HIV/AIDS. These results may reflect adaptations of southern African populations to infectious diseases. Consistent with previous studies, this study highlights the significance of the San in the genetics of human populations, as they are distinct from the other populations in many respects i.e. haplotype structure, locations of recombination hotspots, copy number and population structure. Furthermore, our study demonstrates the admixture of the San, Bantu-speaking populations and populations of Eurasian ancestry in some of the southern and eastern African populations. It illustrates the value in correcting for this stratification in future genome-wide association studies, and suggests that a future admixture mapping in these populations would likely be warranted and successful.

Estimates of the number of human genes that produce more than one transcript isoform through alternative mRNA splicing depend on the assumption that the observation of multiple transcripts from a gene can be attributed entirely to alternative splicing. It is possible, however, that a substantial proportion of cases where multiple transcripts have been observed for a gene result from differences between alleles. Many examples of genes that are spliced differently from different alleles have been reported but no systematic estimate of the proportion of alternatively spliced genes that are affected by such polymorphisms has been carried out. We find that alternative transcript isoforms are non-randomly associated with closely linked nucleotide polymorphisms, based on an integrated analysis of the dbSNP, dbEST and ASAP databases. From the observed level of association between transcript isoforms and polymorphisms, we estimate that 21% of alternatively spliced genes are affected by polymorphisms that either completely determine which form of the transcript is observed or alter the relative abundances of some of the alternative isoforms. We provide a conservative lower bound of 6% on this estimate and point out that alternative splicing cannot be confirmed absolutely unless more than one transcript is observed from the same allele.

We have performed a large-scale analysis of amino acid sequence evolution after gene duplication by comparing evolution after gene duplication with evolution after speciation in over 1,800 phylogenetic trees constructed from manually curated alignments of protein domains downloaded from the PFAM database. The site-specific rate of evolution is significantly altered by gene duplication. A significant increase in the proportion of amino acid substitutions at constrained (slowly evolving) sites after duplication was observed. An increase in the proportion of replacements at normally constrained amino acid sites could result from relaxation of purifying selective pressure. However, the proportion of amino acid replacements involving radical changes in amino acid properties after duplication does not appear to be significantly increased by relaxed selective pressure. The increased proportion of replacements at constrained sites was observed over a relatively large range of protein change (up to 25% amino acid replacements per site). These findings have implications for our understanding of the nature of evolution after duplication and may help to shed light on the evolution of novel protein functions through gene duplication.

Using a large set of orthologous human and mouse gene pairs, we have characterized genes that have been retained in duplicate in human over timescales comparable to the time of speciation of human and mouse. Orthologous gene pairs for which a paralogous gene has been present for much or all of the time since speciation show an increased rate of nonsynonymous substitution. We have related rate of divergence to functional classification using the Gene Ontology terms. Protein function was found, in some cases, to have a larger impact on rate of evolution than the presence or absence of a paralog. No evidence was found that genes that have been retained in duplicate are weighted toward any functional categories. An increase in the ratio of nonsynonymous to synonymous changes following duplication has previously been reported. However, because amino acid sequences include conservative as well as more freely evolving sites, the ratio of nonsynonymous to synonymous changes tends to be higher for closely related pairs. By measuring the divergence of orthologs only and comparing between genes for which a paralogous gene is either present or absent, we have compared gene pairs that share a common divergence time. We have also found that shorter genes have a higher probability of being found duplicated in the human genome, possibly reflecting a mutational effect.

Background Identification of differentially expressed genes from transcriptomic studies is one of the most common mechanisms to identify tumor biomarkers. This approach however is not well suited to identify interaction between genes whose protein products potentially influence each other, which limits its power to identify molecular wiring of tumour cells dictating response to a drug. Due to the fact that signal transduction pathways are not linear and highly interlinked, the biological response they drive may be better described by the relative amount of their components and their functional relationships than by their individual, absolute expression. Results Gene expression microarray data for 109 tumor cell lines with known sensitivity to the death ligand cytokine tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) was used to identify genes with potential functional relationships determining responsiveness to TRAIL-induced apoptosis. The machine learning technique Random Forest in the statistical environment “R” with backward elimination was used to identify the key predictors of TRAIL sensitivity and differentially expressed genes were identified using the software GeneSpring. Gene co-regulation and statistical interaction was assessed with q-order partial correlation analysis and non-rejection rate. Biological (functional) interactions amongst the co-acting genes were studied with Ingenuity network analysis. Prediction accuracy was assessed by calculating the area under the receiver operator curve using an independent dataset. We show that the gene panel identified could predict TRAIL-sensitivity with a very high degree of sensitivity and specificity (AUC = 0 · 84). The genes in the panel are co-regulated and at least 40% of them functionally interact in signal transduction pathways that regulate cell death and cell survival, cellular differentiation and morphogenesis. Importantly, only 12% of the TRAIL-predictor genes were differentially expressed highlighting the importance of functional interactions in predicting the biological response. Conclusions The advantage of co-acting gene clusters is that this analysis does not depend on differential expression and is able to incorporate direct- and indirect gene interactions as well as tissue- and cell-specific characteristics. This approach (1) identified a descriptor of TRAIL sensitivity which performs significantly better as a predictor of TRAIL sensitivity than any previously reported gene signatures, (2) identified potential novel regulators of TRAIL-responsiveness and (3) provided a systematic view highlighting fundamental differences between the molecular wiring of sensitive and resistant cell types. Electronic supplementary material The online version of this article (doi:10.1186/1471-2164-15-1144) contains supplementary material, which is available to authorized users.

Background Substantial progress has been made in the identification of sequence elements that control mRNA splicing and the genetic variants in these elements that alter mRNA splicing (referred to as splicing quantitative trait loci – sQTLs). Genetic variants that affect mRNA splicing in trans are harder to identify because their effects can be more subtle and diffuse, and the variants are not co-located with their targets. We carried out a transcriptome-wide analysis of the effects of a mutation in a ubiquitous splicing factor that causes retinitis pigmentosa (RP) on mRNA splicing, using exon microarrays. Results Exon microarray data was generated from whole blood samples obtained from four individuals with a mutation in the splicing factor PRPF8 and four sibling controls. Although the mutation has no known phenotype in blood, there was evidence of widespread differences in splicing between cases and controls (affecting approximately 20% of exons). Most probesets with significantly different inclusion (defined as the expression intensity of the exon divided by the expression of the corresponding transcript) between cases and controls had higher inclusion in cases and corresponded to exons that were shorter than average, rich, located towards the 5’ end of the gene and flanked by long introns. Introns flanking affected probesets were particularly depleted for the shortest category of introns, associated with splicing via intron definition. Conclusions Our results show that a mutation in a splicing factor, with a phenotype that is restricted to retinal tissue, acts as a trans-sQTL cluster in whole blood samples. Characteristics of the affected exons suggest that they are spliced co-transcriptionally and via exon definition. However, due to the small sample size available for this study, further studies are required to confirm the widespread impact of this PRPF8 mutation on mRNA splicing outside the retina.

Completion of the human genome sequence provides evidence for a gene count with lower bound 30,000–40,000. Significant protein complexity may derive in part from multiple transcript isoforms. Recent EST based studies have revealed that alternate transcription, including alternative splicing, polyadenylation and transcription start sites, occurs within at least 30–40% of human genes. Transcript form surveys have yet to integrate the genomic context, expression, frequency, and contribution to protein diversity of isoform variation. We determine here the degree to which protein coding diversity may be influenced by alternate expression of transcripts by exhaustive manual confirmation of genome sequence annotation, and comparison to available transcript data to accurately associate skipped exon isoforms with genomic sequence. Relative expression levels of transcripts are estimated from EST database representation. The rigorous in silico method accurately identifies exon skipping using verified genome sequence. 545 genes have been studied in this first hand-curated assessment of exon skipping on chromosome 22. Combining manual assessment with software screening of exon boundaries provides a highly accurate and internally consistent indication of skipping frequency. 57 of 62 exon skipping events occur in the protein coding regions of 52 genes. A single gene, (FBXO7) expresses an exon repetition. 59% of highly represented multi-exon genes are likely to express exon-skipped isoforms in ratios that vary from 1:1 to 1:>100. The proportion of all transcripts corresponding to multi-exon genes that exhibit an exon skip is estimated to be 5%.

The short arms of the five acrocentric human chromosomes harbor sequences that direct the assembly and function of the nucleolus, one of the key functional domains of the nucleus, yet they are absent from the current human genome assembly. Here we describe the genomic architecture of these human nucleolar organizers. Sequences distal and proximal to ribosomal gene arrays are conserved among the acrocentric chromosomes, suggesting they are sites of frequent recombination. Although previously believed to be heterochromatic, characterization of these two flanking regions reveals that they share a complex genomic architecture similar to other euchromatic regions of the genome, but they have distinct genomic characteristics. Proximal sequences are almost entirely segmentally duplicated, similar to the regions bordering centromeres. In contrast, the distal sequence is predominantly unique to the acrocentric short arms and is dominated by a very large inverted repeat. We show that the distal element is localized to the periphery of the nucleolus, where it appears to anchor the ribosomal gene repeats. This, combined with its complex chromatin structure and transcriptional activity, suggests that this region is involved in nucleolar organization. Our results provide a platform for investigating the role of NORs in nucleolar formation and function, and open the door for determining the role of these regions in the well-known empirical association of nucleoli with pathology.

Motivation: DNA methylation is an epigenetic mark that can stably repress gene expression. Because of its biological and clinical significance, several methods have been developed to compare genome-wide patterns of methylation between groups of samples. The application of gene set analysis to identify relevant groups of genes that are enriched for differentially methylated genes is often a major component of the analysis of these data. This can be used, for example, to identify processes or pathways that are perturbed in disease development. We show that gene-set analysis, as it is typically applied to genome-wide methylation assays, is severely biased as a result of differences in the numbers of CpG sites associated with different classes of genes and gene promoters. Results: We demonstrate this bias using published data from a study of differential CpG island methylation in lung cancer and a dataset we generated to study methylation changes in patients with long-standing ulcerative colitis. We show that several of the gene sets that seem enriched would also be identified with randomized data. We suggest two existing approaches that can be adapted to correct the bias. Accounting for the bias in the lung cancer and ulcerative colitis datasets provides novel biological insights into the role of methylation in cancer development and chronic inflammation, respectively. Our results have significant implications for many previous genome-wide methylation studies that have drawn conclusions on the basis of such strongly biased analysis. Contact: cathal.seoighe@nuigalway.ie Supplementary information: Supplementary data are available at Bioinformatics online.

Background microRNAs (miRNAs) have been shown to regulate the expression of a large number of genes and play key roles in many biological processes. Several previous studies have quantified the inhibitory effect of a miRNA indirectly by considering the expression levels of genes that are predicted to be targeted by the miRNA and this approach has been shown to be robust to the choice of prediction algorithm. Given a gene expression dataset, Cheng et al. defined the regulatory effect score (RE-score) of a miRNA as the difference in the gene expression rank of targets of the miRNA compared to non-targeted genes. Results Using microarray data from parent-offspring trios from the International HapMap project, we show that the RE-score of most miRNAs is correlated between parents and offspring and, thus, inter-individual variation in RE-score has a genetic component in humans. Indeed, the mean RE-score across miRNAs is correlated between parents and offspring, suggesting genetic differences in the overall efficiency of the miRNA biogenesis pathway between individuals. To explore the genetics of this quantitative trait further, we carried out a genome-wide association study of the mean RE-score separately in two HapMap populations (CEU and YRI). No genome-wide significant associations were discovered; however, a SNP rs17409624, in an intron of DROSHA, was significantly associated with mean RE-score in the CEU population following permutation-based control for multiple testing based on all SNPs mapped to the canonical miRNA biogenesis pathway; of 244 individual miRNA RE-scores assessed in the CEU, 214 were associated (p < 0.05) with rs17409624. The SNP was also nominally significantly associated (p = 0.04) with mean RE-score in the YRI population. Interestingly, the same SNP was associated with 17 (8.5% of all expressed) miRNA expression levels in the CEU. We also show here that the expression of the targets of most miRNAs is more highly correlated with global changes in miRNA regulatory effect than with the expression of the miRNA itself. Conclusions We present evidence that miRNA regulatory effect is a heritable trait in humans and that a polymorphism of the DROSHA gene contributes to the observed inter-individual differences.

Background We carried out an analysis of intron length conservation across a diverse group of nineteen mammalian species. Motivated by recent research suggesting a role for time delays associated with intron transcription in gene expression oscillations required for early embryonic patterning, we searched for examples of genes that showed the most extreme conservation of total intron content in mammals. Results Gene sets annotated as being involved in pattern specification in the early embryo or containing the homeobox DNA-binding domain, were significantly enriched among genes with highly conserved intron content. We used ancestral sequences reconstructed with probabilistic models that account for insertion and deletion mutations to distinguish insertion and deletion events on lineages leading to human and mouse from their last common ancestor. Using a randomization procedure, we show that genes containing the homeobox domain show less change in intron content than expected, given the number of insertion and deletion events within their introns. Conclusions Our results suggest selection for gene expression precision or the existence of additional development-associated genes for which transcriptional delay is functionally significant.

Heterogeneity in sample composition is an inherent issue in many gene expression studies and, in many cases, should be taken into account in the downstream analysis to enable correct interpretation of the underlying biological processes. Typical examples are infectious diseases or immunology-related studies using blood samples, where, for example, the proportions of lymphocyte sub-populations are expected to vary between cases and controls. Nonnegative Matrix Factorization (NMF) is an unsupervised learning technique that has been applied successfully in several fields, notably in bioinformatics where its ability to extract meaningful information from high-dimensional data such as gene expression microarrays has been demonstrated. Very recently, it has been applied to biomarker discovery and gene expression deconvolution in heterogeneous tissue samples. Being essentially unsupervised, standard NMF methods are not guaranteed to find components corresponding to the cell types of interest in the sample, which may jeopardize the correct estimation of cell proportions. We have investigated the use of prior knowledge, in the form of a set of marker genes, to improve gene expression deconvolution with NMF algorithms. We found that this improves the consistency with which both cell type proportions and cell type gene expression signatures are estimated. The proposed method was tested on a microarray dataset consisting of pure cell types mixed in known proportions. Pearson correlation coefficients between true and estimated cell type proportions improved substantially (typically from about 0.5 to approximately 0.8) with the semi-supervised (marker-guided) versions of commonly used NMF algorithms. Furthermore known marker genes associated with each cell type were assigned to the correct cell type more frequently for the guided versions. We conclude that the use of marker genes improves the accuracy of gene expression deconvolution using NMF and suggest modifications to how the marker gene information is used that may lead to further improvements.

Small noncoding RNAs regulate a variety of cellular processes, including genomic imprinting, chromatin remodeling, replication, transcription, and translation. Here, we report small replication-regulating RNAs (srRNAs) that specifically inhibit DNA replication of the human BK polyomavirus (BKV) in vitro and in vivo . srRNAs from FM3A murine mammary tumor cells were enriched by DNA replication assay-guided fractionation and hybridization to the BKV noncoding control region (NCCR) and synthesized as cDNAs. Selective mutagenesis of the cDNA sequences and their putative targets suggests that the inhibition of BKV DNA replication is mediated by srRNAs binding to the viral NCCR, hindering early steps in the initiation of DNA replication. Ectopic expression of srRNAs in human cells inhibited BKV DNA replication in vivo . Additional srRNAs were designed and synthesized that specifically inhibit simian virus 40 (SV40) DNA replication in vitro . These observations point to novel mechanisms for regulating DNA replication and suggest the design of synthetic agents for inhibiting replication of polyomaviruses and possibly other viruses.

The RV144 vaccine trial in Thailand demonstrated that an HIV vaccine could prevent infection in humans and highlights the importance of understanding protective immunity against HIV. We used a nonhuman primate model to define immune and genetic mechanisms of protection against mucosal infection by the simian immunodeficiency virus (SIV). A plasmid DNA prime/recombinant adenovirus serotype 5 (rAd5) boost vaccine regimen was evaluated for its ability to protect monkeys from infection by SIVmac251 or SIVsmE660 isolates after repeat intrarectal challenges. Although this prime-boost vaccine regimen failed to protect against SIVmac251 infection, 50% of vaccinated monkeys were protected from infection with SIVsmE660. Among SIVsmE660-infected animals, there was about a one-log reduction in peak plasma virus RNA in monkeys expressing the major histocompatibility complex class I allele Mamu-A*01, implicating cytotoxic T lymphocytes in the control of SIV replication once infection is established. Among Mamu-A*01-negative monkeys challenged with SIVsmE660, no CD8(+) T cell response or innate immune response was associated with protection against virus acquisition. However, low levels of neutralizing antibodies and an envelope-specific CD4(+) T cell response were associated with vaccine protection in these monkeys. Moreover, monkeys that expressed two TRIM5 alleles that restrict SIV replication were more likely to be protected from infection than monkeys that expressed at least one permissive TRIM5 allele. This study begins to elucidate the mechanisms of vaccine protection against immunodeficiency viruses and highlights the need to analyze these immune and genetic correlates of protection in future trials of HIV vaccine strategies.

Background Nonnegative Matrix Factorization (NMF) is an unsupervised learning technique that has been applied successfully in several fields, including signal processing, face recognition and text mining. Recent applications of NMF in bioinformatics have demonstrated its ability to extract meaningful information from high-dimensional data such as gene expression microarrays. Developments in NMF theory and applications have resulted in a variety of algorithms and methods. However, most NMF implementations have been on commercial platforms, while those that are freely available typically require programming skills. This limits their use by the wider research community. Results Our objective is to provide the bioinformatics community with an open-source, easy-to-use and unified interface to standard NMF algorithms, as well as with a simple framework to help implement and test new NMF methods. For that purpose, we have developed a package for the R/BioConductor platform. The package ports public code to R, and is structured to enable users to easily modify and/or add algorithms. It includes a number of published NMF algorithms and initialization methods and facilitates the combination of these to produce new NMF strategies. Commonly used benchmark data and visualization methods are provided to help in the comparison and interpretation of the results. Conclusions The NMF package helps realize the potential of Nonnegative Matrix Factorization, especially in bioinformatics, providing easy access to methods that have already yielded new insights in many applications. Documentation, source code and sample data are available from CRAN.

Background In eukaryotes mRNA transcripts of protein-coding genes in which an intron has been retained in the coding region normally result in premature stop codons and are therefore degraded through the nonsense-mediated mRNA decay (NMD) pathway. There is evidence in the form of selective pressure for in-frame stop codons in introns and a depletion of length three introns that this is an important and conserved quality-control mechanism. Yet recent reports have revealed that the efficiency of NMD varies across tissues and between individuals, with important clinical consequences. Principal Findings Using previously published Affymetrix exon microarray data from cell lines genotyped as part of the International HapMap project, we investigated whether there are heritable, inter-individual differences in the abundance of intron-containing transcripts, potentially reflecting differences in the efficiency of NMD. We identified intronic probesets using EST data and report evidence of heritability in the extent of intron expression in 56 HapMap trios. We also used a genome-wide association approach to identify genetic markers associated with intron expression. Among the top candidates was a SNP in the DCP1A gene, which forms part of the decapping complex, involved in NMD. Conclusions While we caution that some of the apparent inter-individual difference in intron expression may be attributable to different handling or treatments of cell lines, we hypothesize that there is significant polymorphism in the process of NMD, resulting in heritable differences in the abundance of intronic mRNA. Part of this phenotype is likely to be due to a polymorphism in a decapping enzyme on human chromosome 3.

The objective of the study was to examine the association between a functional 4 bp proinsulin gene insertion polymorphism (IVS-69), fasting insulin concentrations, and body composition in black South African women. Body composition, body fat distribution, fasting glucose and insulin concentrations, and IVS-69 genotype were measured in 115 normal-weight (BMI < 25 kg/ m(2)) and 138 obese (BMI >= 30 kg/m(2)) premenopausal women. The frequency of the insertion allele was significantly higher in the class 2 obese (BMI >= 35kg/m(2)) compared with the normal-weight group (P = 0.029). Obese subjects with the insertion allele had greater fat mass (42.3 +/- 0.9 vs. 38.9 +/- 0.9 kg, P = 0.034) and fat-free soft tissue mass (47.4 +/- 0.6 vs. 45.1 +/- 0.6 kg, P = 0.014), and more abdominal subcutaneous adipose tissue (SAT, 595 +/- 17 vs. 531 +/- 17 cm(2), P = 0.025) but not visceral fat (P = 0.739), than obese homozygotes for the wild-type allele. Only SAT was greater in normal-weight subjects with the insertion allele (P = 0.048). There were no differences in fasting insulin or glucose levels between subjects with the insertion allele or homozygotes for the wild-type allele in the normal-weight or obese groups. In conclusion, the 4 bp proinsulin gene insertion allele is associated with extreme obesity, reflected by greater fat-free soft tissue mass and fat mass, particularly SAT, in obese black South African women.

Identifying the specific genetic characteristics of successfully transmitted variants may prove central to the development of effective vaccine and microbicide interventions. Although human immunodeficiency virus transmission is associated with a population bottleneck, the extent to which different factors influence the diversity of transmitted viruses is unclear. We estimate here the number of transmitted variants in 69 heterosexual men and women with primary subtype C infections. From 1,505 env sequences obtained using a single genome amplification approach we show that 78% of infections involved single variant transmission and 22% involved multiple variant transmissions (median of 3). We found evidence for mutations selected for cytotoxic-T-lymphocyte or antibody escape and a high prevalence of recombination in individuals infected with multiple variants representing another potential escape pathway in these individuals. In a combined analysis of 171 subtype B and C transmission events, we found that infection with more than one variant does not follow a Poisson distribution, indicating that transmission of individual virions cannot be seen as independent events, each occurring with low probability. While most transmissions resulted from a single infectious unit, multiple variant transmissions represent a significant fraction of transmission events, suggesting that there may be important mechanistic differences between these groups that are not yet understood.

Background The cytotoxic T-lymphocyte immune response is important in controlling HIV-1 replication in infected humans. In this immune pathway, viral peptides within infected cells are presented to T-lymphocytes by the polymorphic human leukocyte antigens (HLA). HLA alleles exert selective pressure on the peptide regions and immune escape mutations that occur at some of the targeted sites can enable the virus to adapt to the infected host. The pattern of ongoing immune escape and reversion associated with several human HLA alleles has been studied extensively. Such mutations revert upon transmission to a host without the HLA allele because the escape mutation incurs a fitness cost. However, to-date there has been little attempt to study permanent loss of CTL epitopes due to escape mutations without an effect on fitness. Results Here, we set out to determine the extent of adaptation of HIV-1 to three well-characterized HLA alleles during the initial exposure of the virus to the human cytotoxic immune responses following transmission from chimpanzee. We generated a chimpanzee consensus sequence to approximate the virus sequence that was initially transmitted to the human host and used a method based on peptide binding affinity to HLA crystal structures to predict peptides that were potentially targeted by the HLA alleles on this sequence. Next, we used codon-based phylogenetic models to quantify the average selective pressure that acted on these regions during the period immediately following the zoonosis event, corresponding to the branch of the phylogenetic tree leading to the common ancestor of all of the HIV-1 sequences. Evidence for adaptive evolution during this period was observed at regions recognised by HLA A*6801 and A*0201, both of which are common in African populations. No evidence of adaptive evolution was observed at sites targeted by HLA-B*2705, which is a rare allele in African populations. Conclusion Our results suggest that the ancestral HIV-1 virus experienced a period of positive selective pressure due to immune responses associated with HLA alleles that were common in the infected human population. We propose that this resulted in permanent escape from immune responses targeting unconstrained regions of the virus.

Host immune responses against infectious pathogens exert strong selective pressures favouring the emergence of escape mutations that prevent immune recognition. Escape mutations within or flanking functionally conserved epitopes can occur at a significant cost to the pathogen in terms of its ability to replicate effectively. Such mutations come under selective pressure to revert to the wild type in hosts that do not mount an immune response against the epitope. Amino acid positions exhibiting this pattern of escape and reversion are of interest because they tend to coincide with immune responses that control pathogen replication effectively. We have used a probabilistic model of protein coding sequence evolution to detect sites in HIV-1 exhibiting a pattern of rapid escape and reversion. Our model is designed to detect sites that toggle between a wild type amino acid, which is susceptible to a specific immune response, and amino acids with lower replicative fitness that evade immune recognition. Through simulation, we show that this model has significantly greater power to detect selection involving immune escape and reversion than standard models of diversifying selection, which are sensitive to an overall increased rate of non-synonymous substitution. Applied to alignments of HIV-1 protein coding sequences, the model of immune escape and reversion detects a significantly greater number of adaptively evolving sites in env and nef. In all genes tested, the model provides a significantly better description of adaptively evolving sites than standard models of diversifying selection. Several of the sites detected are corroborated by association between Human Leukocyte Antigen (HLA) and viral sequence polymorphisms. Overall, there is evidence for a large number of sites in HIV-1 evolving under strong selective pressure, but exhibiting low sequence diversity. A phylogenetic model designed to detect rapid toggling between wild type and escape amino acids identifies a larger number of adaptively evolving sites in HIV-1, and can in some cases correctly identify the amino acid that is susceptible to the immune response., Author Summary Viruses, such as HIV, are able to evade host immune responses through escape mutations, yet sometimes they do so at a cost. This cost is the reduction in the ability of the virus to replicate, and thus selective pressure exists for a virus to revert to its original state in the absence of the host immune response that caused the initial escape mutation. This pattern of escape and reversion typically occurs when viruses are transmitted between individuals with different immune responses. We develop a phylogenetic model of immune escape and reversion and provide evidence that it outperforms existing models for the detection of selective pressure associated with host immune responses. Finally, we demonstrate that amino acid toggling is a pervasive process in HIV-1 evolution, such that many of the positions in the virus that evolve rapidly, under the influence of positive Darwinian selection, nonetheless display quite low sequence diversity. This highlights the limitations of HIV-1 evolution, and sites such as these are potentially good targets for HIV-1 vaccines.

One of the most important genetic factors known to affect the rate of disease progression in HIV-infected individuals is the genotype at the Class I Human Leukocyte Antigen (HLA) locus, which determines the HIV peptides targeted by cytotoxic T-lymphocytes (CTLs). Individuals with HLA-B*57 or B*5801 alleles, for example, target functionally important parts of the Gag protein. Mutants that escape these CTL responses may have lower fitness than the wild-type and can be associated with slower disease progression. Transmission of the escape variant to individuals without these HLA alleles is associated with rapid reversion to wild-type. However, the question of whether infection with an escape mutant offers an advantage to newly infected hosts has not been addressed. Here we investigate the relationship between the genotypes of transmitted viruses and prognostic markers of disease progression and show that infection with HLA-B*57/B*5801 escape mutants is associated with lower viral load and higher CD4+ counts., Author Summary Following infection with HIV, it is well established that a person's genetic makeup is a major determinant of how quickly they will progress to AIDS. Particularly important is the class I Human leukocyte antigen (HLA) gene that is responsible for alerting the immune system to HIV's presence. One of the reasons our immune systems are unable to beat HIV is that the virus can mutate to forms that our HLA genes no longer recognise. However, some people have versions of the HLA gene (for example HLA-B*57 and HLA-B*5801) that are known to force HIV to tolerate mutations that damage its ability to reproduce. Slower HIV reproduction is thought to be one reason that HLA-B*57 and HLA-B*5801 positive people progress to AIDS more slowly than most other HIV infected persons. We report here on a study of HLA-B*57 and HLA-B*5801 negative women in which better control of disease tended to be associated with their being infected with viruses carrying mutations that have been previously shown to reduce replication. These mutations characterise viruses found infecting HLA-B*57 and HLA-B*5801 positive people. This indicates for the first time that HLA-B*57 or HLA-B*5801 negative people that are infected by such reproductively compromised viruses may also experience better survival prospects.

Background Accurate mRNA splicing depends on multiple regulatory signals encoded in the transcribed RNA sequence. Many examples of mutations within human splice regulatory regions that alter splicing qualitatively or quantitatively have been reported and allelic differences in mRNA splicing are likely to be a common and important source of phenotypic diversity at the molecular level, in addition to their contribution to genetic disease susceptibility. However, because the effect of a mutation on the efficiency of mRNA splicing is often difficult to predict, many mutations that cause disease through an effect on splicing are likely to remain undiscovered. Results We have combined a genome-wide scan for sequence polymorphisms likely to affect mRNA splicing with analysis of publicly available Expressed Sequence Tag (EST) and exon array data. The genome-wide scan uses published tools and identified 30,977 SNPs located within donor and acceptor splice sites, branch points and exonic splicing enhancer elements. For 1,185 candidate splicing polymorphisms the difference in splicing between alternative alleles was corroborated by publicly available exon array data from 166 lymphoblastoid cell lines. We developed a novel probabilistic method to infer allele-specific splicing from EST data. The method uses SNPs and alternative mRNA isoforms mapped to EST sequences and models both regulated alternative splicing as well as allele-specific splicing. We have also estimated heritability of splicing and report that a greater proportion of genes show evidence of splicing heritability than show heritability of overall gene expression level. Our results provide an extensive resource that can be used to assess the possible effect on splicing of human polymorphisms in putative splice-regulatory sites. Conclusion We report a set of genes showing evidence of allele-specific splicing from an integrated analysis of genomic polymorphisms, EST data and exon array data, including several examples for which there is experimental evidence of polymorphisms affecting splicing in the literature. We also present a set of novel allele-specific splicing candidates and discuss the strengths and weaknesses of alternative technologies for inferring the effect of sequence variants on mRNA splicing.

Cytotoxic T lymphocyte (CTL) responses are thought to be essential for the control of HIV-1 replication in vivo and immunogens that elicit CTL responses are currently a major focus of HIV vaccine research. Here we investigated two aspects of the CTL response to HIV-1 subtype C that are important for vaccine design and efficacy monitoring. First, we assessed the relationship between the CTL response and sequence diversity, using a robust statistical method. While peptides that were most frequently recognized by the CTL response in Nef and p24 tended to be conserved, this was not the case for p17 where epitope recognition coincided with highly variable regions. Second, we investigated the relationship between observed and predicted CTL responses, given the HLA genotype of infected individuals. Only 52% of the Nef peptides and 64% of the Gag peptides that elicited a CTL response contained sequence motifs thought to be required for binding by the HLA-A or -B alleles found in the corresponding patient. In a comparable subtype B dataset a much higher proportion of the peptides that elicited a CTL response were consistent with the patient HLA genotype (96% and 83% for Nef and Gag, respectively). We demonstrate that this difference between subtypes C and B is likely to result from a combination of a tendency for HLA alleles common in Southern African populations to be poorly characterized, as well as a tendency for sequence motifs associated with HLA recognition to be overspecified for sequence variation found in the B clade. Our results suggest that knowledge of HLA binding motifs is likely to be biased toward certain populations and subtypes. This can have important implications for understanding immune escape and predicting vaccine efficacy in the context of populations primarily infected with non-B subtype HIV-1.

Understanding how pathogens acquire resistance to drugs is important for the design of treatment strategies, particularly for rapidly evolving viruses such as HIV-1. Drug treatment can exert strong selective pressures and sites within targeted genes that confer resistance frequently evolve far more rapidly than the neutral rate. Rapid evolution at sites that confer resistance to drugs can be used to help elucidate the mechanisms of evolution of drug resistance and to discover or corroborate novel resistance mutations. We have implemented standard maximum likelihood methods that are used to detect diversifying selection and adapted them for use with serially sampled reverse transcriptase (RT) coding sequences isolated from a group of 300 HIV-1 subtype C-infected women before and after single-dose nevirapine (sdNVP) to prevent mother-to-child transmission. We have also extended the standard models of codon evolution for application to the detection of directional selection. Through simulation, we show that the directional selection model can provide a substantial improvement in sensitivity over models of diversifying selection. Five of the sites within the RT gene that are known to harbor mutations that confer resistance to nevirapine (NVP) strongly supported the directional selection model. There was no evidence that other mutations that are known to confer NVP resistance were selected in this cohort. The directional selection model, applied to serially sampled sequences, also had more power than the diversifying selection model to detect selection resulting from factors other than drug resistance. Because inference of selection from serial samples is unlikely to be adversely affected by recombination, the methods we describe may have general applicability to the analysis of positive selection affecting recombining coding sequences when serially sampled data are available.

Comparative analysis of orthologues from diverse vertebrates can be used to identify molecular signatures that are important for gene function and which may predict novel regulatory mechanisms or explain morphological diversity. The forkhead box G1 (FoxG1) transcription factor is potentially a strong candidate gene for determining forebrain size in vertebrates due to its role in the development of the telencephalon, where it promotes progenitor proliferation and suppresses premature neurogenesis. To investigate the role of FoxG1 in forebrain evolution, we cloned and analyzed the cDNA sequences for nine new FoxG1 orthologues, including six mammals and three reptiles, and show that there is an extended proline and glutamine region in the N-terminal domain that is specific to mammals. In contrast to some previous studies of other potential determinants of brain size, we find no evidence that the coding sequence of FoxG1 has evolved under positive selection in vertebrates. Previously published work has indicated that FOXG1 was duplicated in humans, and two forms, FOXG1A and FOXG1B, are present in the Entrez Gene database. We report that FOXG1 has not been duplicated in humans and that FOXG1A is likely to be an artifact. Our comparative analysis of FOXG1B and its orthologues has revealed a very high level of conservation in the 3' untranslated region (UTR). Using available computational tools, we find evidence for conserved recognition sites for the miR-9 and miR-33 microRNAs in the FoxG1 3' UTR and hypothesize that these brain-expressed microRNAs may regulate FoxG1 post-transcriptionally during forebrain development.

Summary: Gene expression data are typically generated from heterogeneous biological samples that are composed of multiple cell or tissue types, in varying proportions, each contributing to global gene expression. This heterogeneity is a major confounder in standard analysis such as differential expression analysis, where differences in the relative proportions of the constituent cells may prevent or bias the detection of cell-specific differences. Computational deconvolution of global gene expression is an appealing alternative to costly physical sample separation techniques and enables a more detailed analysis of the underlying biological processes at the cell-type level. To facilitate and popularize the application of such methods, we developed CellMix, an R package that incorporates most state-of-the-art deconvolution methods, into an intuitive and extendible framework, providing a single entry point to explore, assess and disentangle gene expression data from heterogeneous samples. Availability and Implementation: The CellMix package builds on R/BioConductor and is available from http://web.cbio.uct.ac.za/∼renaud/CRAN/web/CellMix. It is currently being submitted to BioConductor. The package’s vignettes notably contain additional information, examples and references. Contact: renaud@cbio.uct.ac.za

This longitudinal study aimed to determine the incidence and pathogenic implications of dual human immunodeficiency virus type 1 (HIV-1) infection in a cohort of female sex workers. Blood samples from 31 recently infected women were screened by use of a heteroduplex mobility assay and sequencing. The median viral load set point was 5404 copies/ mL (n = 22), which was measured by use of the bDNA assay. Within 3 months of infection, 19% (6/31) of the women were dually infected with 2 distinct HIV-1 subtype C viruses. No evidence of superinfection was detected over the course of 24 months of follow-up, indicating that the risk of dual infection is highest around the time of the initial infection. There was a significant association between dual infection and elevated viral load set point.

Background Rates of substitution in protein-coding sequences can provide important insights into evolutionary processes that are of biomedical and theoretical interest. Increased availability of coding sequence data has enabled researchers to estimate more accurately the coding sequence divergence of pairs of organisms. However the use of different data sources, alignment protocols and methods to estimate substitution rates leads to widely varying estimates of key parameters that define the coding sequence divergence of orthologous genes. Although complete genome sequence data are not available for all organisms, fragmentary sequence data can provide accurate estimates of substitution rates provided that an appropriate and consistent methodology is used and that differences in the estimates obtainable from different data sources are taken into account. Results We have developed FRAGS, an application framework that uses existing, freely available software components to construct in-frame alignments and estimate coding substitution rates from fragmentary sequence data. Coding sequence substitution estimates for human and chimpanzee sequences, generated by FRAGS, reveal that methodological differences can give rise to significantly different estimates of important substitution parameters. The estimated substitution rates were also used to infer upper-bounds on the amount of sequencing error in the datasets that we have analysed. Conclusion We have developed a system that performs robust estimation of substitution rates for orthologous sequences from a pair of organisms. Our system can be used when fragmentary genomic or transcript data is available from one of the organisms and the other is a completely sequenced genome within the Ensembl database. As well as estimating substitution statistics our system enables the user to manage and query alignment and substitution data.

Genetic variation in gene expression makes an important contribution to phenotypic variation and susceptibility to disease. Recently, a subset of cis -acting expression quantitative loci (eQTLs) has been found to result from polymorphisms that affect RNA stability. Here we carried out a search for trans -acting variants that influence RNA stability. We first demonstrate that differences in the activity of trans -acting factors that stabilize RNA can be detected by comparing the expression levels of long-lived (stable) and short-lived (unstable) transcripts in high-throughput gene expression experiments. Using gene expression microarray data generated from eight HapMap3 populations, we calculated the relative expression ranks of long-lived transcripts versus short-lived transcripts in each sample. Treating this as a quantitative trait, we applied genome-wide association and identified a single nucleotide polymorphism (SNP), rs6137010, on chromosome 20p13 with which it is strongly associated in two Asian populations ( p = 4×10 −10 in CHB - Han Chinese from Beijing; p = 1×10 −4 in JPT - Japanese from Tokyo). This SNP is a cis -eQTL for SNRPB in CHB and JPT but not in the other six HapMap3 populations. SNRPB is a core component of the spliceosome, and has previously been shown to affect the expression of many RNA processing factors. We propose that a cis -eQTL of SNRPB may be directly responsible for inter-individual variation in relative expression of long-lived versus short-lived transcript in Asian populations. In support of this hypothesis, knockdown of SNRPB results in a significant reduction in the relative expression of long-lived versus short-lived transcripts. Samples with higher relative expression of long-lived transcripts also had higher relative expression of coding compared to non-coding RNA and of RNA from housekeeping compared to non-housekeeping genes, due to the lower decay rates of coding RNAs, particularly those that perform housekeeping functions, compared to non-coding RNAs.

The major histocompatibility (MHC) molecules are capable of presenting neoantigens resulting from somatic mutations on cell surfaces, potentially directing immune responses against cancer. This led to the hypothesis that cancer driver mutations may occur in gaps in the capacity to present neoantigens that are dependent on MHC genotype. If this is correct, it has important implications for understanding oncogenesis and may help to predict driver mutations based on genotype data. In support of this hypothesis, it has been reported that driver mutations that occur frequently tend to be poorly presented by common MHC alleles and that the capacity of a patient’s MHC alleles to present the resulting neoantigens is predictive of the driver mutations that are observed in their tumour. Here we show that these reports of a strong relationship between driver mutation occurrence and patient MHC alleles are a consequence of unjustified statistical assumptions. Our reanalysis of the data provides no evidence of an effect of MHC genotype on the oncogenic mutation landscape.