Alena Myšičková

Evaluation of methods for modeling transcription factor sequence specificity

Genomic analyses often involve scanning for potential transcription factor (TF) binding sites using models of the sequence specificity of DNA binding proteins. Many approaches have been developed to model and learn a proteins DNA-binding specificity, but these methods have not been systematically compared. Here we applied 26 such approaches to in vitro protein binding microarray data for 66 mouse TFs belonging to various families. For nine TFs, we also scored the resulting motif models on in vivo data, and found that the best in vitro–derived motifs performed similarly to motifs derived from the in vivo data. Our results indicate that simple models based on mononucleotide position weight matrices trained by the best methods perform similarly to more complex models for most TFs examined, but fall short in specific cases (<10% of the TFs examined here). In addition, the best-performing motifs typically have relatively low information content, consistent with widespread degeneracy in eukaryotic TF sequence preferences.

Serum hepatitis B virus RNA levels as an early predictor of hepatitis B envelope antigen seroconversion during treatment with polymerase inhibitors

Hepatitis B envelope antigen (HBeAg) seroconversion represents an endpoint of treatment of chronic hepatitis B virus (HBV) infections. We have studied whether levels of serum HBV RNA during polymerase inhibitor treatment might be helpful for predicting HBeAg seroconversion. HBV RNA levels were determined in serial serum samples from 62 patients with chronic HBV infection (50 HBeAg positive). Patients received antiviral treatment for a mean duration of 30 ± 15 (range, 4‐64) months. A new rapid amplification of complimentary DNA‐ends‐based real‐time polymerase chain reaction was established for quantitative analysis of polyadenylated full‐length (fl) and truncated (tr) HBV RNA. HBV RNA, HBV DNA, and hepatitis B surface antigen (HBsAg) levels as well as presence of HBeAg and hepatitis B envelope antibody were measured at baseline, month 3, month 6, and subsequent time points. Fifteen patients who achieved HBeAg seroconversion after a mean duration of 19 ± 14 (range, 3‐56) months of antiviral treatment showed a significantly stronger decline in mean HBV flRNA and trRNA levels from baseline to month 3 of 1.0 ± 1.4 (range, −1.6‐3.4) and 2.1 ± 1.4 (range, 0‐3.9) and to month 6 of 1.8 ± 1.4 (range, 0‐4.6) and 3.1 ± 1.7 (range, 0‐5.1) log10 copies/mL, respectively, in comparison to 35 HBeAg‐positive patients without HBeAg seroconversion (P < 0.001 for months 3 and 6). A similar decline in HBV RNA levels was observed in HBeAg‐negative patients. The decline of HBV RNA levels at months 3 and 6 of treatment was to be the strongest predictor of HBeAg seroconversion, when compared to levels of HBV DNA, HBsAg, alanine aminotransferase, and HBV genotype, age, and sex. Conclusion: Serum HBV RNA levels may serve as a novel tool for prediction of serological response during polymerase inhibitor treatment in HBeAg‐positive patients. (Hepatology 2015;61:66–76)

Modeling Read Counts for CNV Detection in Exome Sequencing Data

Varying depth of high-throughput sequencing reads along a chromosome makes it possible to observe copy number variants (CNVs) in a sample relative to a reference. In exome and other targeted sequencing projects, technical factors increase variation in read depth while reducing the number of observed locations, adding difficulty to the problem of identifying CNVs. We present a hidden Markov model for detecting CNVs from raw read count data, using background read depth from a control set as well as other positional covariates such as GC-content. The model, exomeCopy, is applied to a large chromosome X exome sequencing project identifying a list of large unique CNVs. CNVs predicted by the model and experimentally validated are then recovered using a cross-platform control set from publicly available exome sequencing data. Simulations show high sensitivity for detecting heterozygous and homozygous CNVs, outperforming normalization and state-of-the-art segmentation methods.

Detection of interacting transcription factors in human tissues using predicted DNA binding affinity

BackgroundTissue-specific gene expression is generally regulated by combinatorial interactions among transcription factors (TFs) which bind to the DNA. Despite this known fact, previous discoveries of the mechanism that controls gene expression usually consider only a single TF.ResultsWe provide a prediction of interacting TFs in 22 human tissues based on their DNA-binding affinity in promoter regions. We analyze all possible pairs of 130 vertebrate TFs from the JASPAR database. First, all human promoter regions are scanned for single TF-DNA binding affinities with TRAP and for each TF a ranked list of all promoters ordered by the binding affinity is created. We then study the similarity of the ranked lists and detect candidates for TF-TF interaction by applying a partial independence test for multiway contingency tables. Our candidates are validated by both known protein-protein interactions (PPIs) and known gene regulation mechanisms in the selected tissue. We find that the known PPIs are significantly enriched in the groups of our predicted TF-TF interactions (2 and 7 times more common than expected by chance). In addition, the predicted interacting TFs for studied tissues (liver, muscle, hematopoietic stem cell) are supported in literature to be active regulators or to be expressed in the corresponding tissue.ConclusionsThe findings from this study indicate that tissue-specific gene expression is regulated by one or two central regulators and a large number of TFs interacting with these central hubs. Our results are in agreement with recent experimental studies.

Stochastic population forecast for Germany and its consequence for the German pension system

Population forecasts are crucial for many social, political and economic decisions. Official population projections rely in general on deterministic models which use different scenarios for future vital rates to indicate uncertainty. However, this technique shows substantial weak points such as assuming absolute correlations between the demographic components. In this paper, we argue that a stochastic projection alternative, with no a priori assumptions provides point forecasts and probabilistic prediction intervals for demographic parameters in addition. Age-sex specific population forecast for Germany is derived through a stochastic population renewal process using forecasts of mortality, fertility and migration. Time series models with demographic restrictions are used to describe immigration, emigration and time varying indices of mortality and fertility rates. These models are then used in the simulation of future vital rates to obtain age-specific population forecast using the cohort-component method. The consequence for the German pension system is discussed. To maintain the actual average pension level the premium rate of the present system rises at least by 50% as the old-age ratio nearly doubles by 2040.

An Inference and Integration Approach for the Consolidation of Ranked Lists

In this article, we describe a new approach that combines the estimation of the lengths of highly conforming sublists with their stochastic aggregation, to deal with two or more rankings of the same set of objects. The goal is to obtain a much smaller set of informative common objects in a new rank order. The input lists can be of large or huge size, their rankings irregular and incomplete due to random and missing assignments. A moderate deviation-based inference procedure and a cross-entropy Monte Carlo technique are used to handle the combinatorial complexity of the task. Two alternative distance measures are considered that can accommodate truncated list information. Finally, the outlined approach is applied to simulated data that was motivated by microarray meta-analysis, an important field of application.

Inferring gene expression from ribosomal promoter sequences, a crowdsourcing approach

The Gene Promoter Expression Prediction challenge consisted of predicting gene expression from promoter sequences in a previously unknown experimentally generated data set. The challenge was presented to the community in the framework of the sixth Dialogue for Reverse Engineering Assessments and Methods (DREAM6), a community effort to evaluate the status of systems biology modeling methodologies. Nucleotide-specific promoter activity was obtained by measuring fluorescence from promoter sequences fused upstream of a gene for yellow fluorescence protein and inserted in the same genomic site of yeast Saccharomyces cerevisiae. Twenty-one teams submitted results predicting the expression levels of 53 different promoters from yeast ribosomal protein genes. Analysis of participant predictions shows that accurate values for low-expressed and mutated promoters were difficult to obtain, although in the latter case, only when the mutation induced a large change in promoter activity compared to the wild-type sequence. As in previous DREAM challenges, we found that aggregation of participant predictions provided robust results, but did not fare better than the three best algorithms. Finally, this study not only provides a benchmark for the assessment of methods predicting activity of a specific set of promoters from their sequence, but it also shows that the top performing algorithm, which used machine-learning approaches, can be improved by the addition of biological features such as transcription factor binding sites.