With the development of high-throughput technologies, genomics datasets rapidly grow in size, including functional genomics data. This has allowed the training of large Deep Learning (DL) models to predict epigenetic readouts, such as protein binding or histone modifications, from genome sequences. However, large dataset sizes come at a price of data consistency, often aggregating results from a large number of studies, conducted under varying experimental conditions. While data from large-scale consortia are useful as they allow studying the effects of different biological conditions, they can also contain unwanted biases from confounding experimental factors. Here, we introduce Metadata-guided Feature Disentanglement (MFD)-an approach that allows disentangling biologically relevant features from potential technical biases. MFD incorporates target metadata into model training, by conditioning weights of the model output layer on different experimental factors. It then separates the factors into disjoint groups and enforces independence of the corresponding feature subspaces with an adversarially learned penalty. We show that the metadata-driven disentanglement approach allows for better model introspection, by connecting latent features to experimental factors, without compromising, or even improving performance in downstream tasks, such as enhancer prediction, or genetic variant discovery. The code will be made available at https://github.com/HealthML/MFD.

Motivation: RNA-binding proteins (RBPs) regulate every aspect of RNA metabolism and function. There are hundreds of RBPs encoded in the eukaryotic genomes, and each recognize its RNA targets through a specific mixture of RNA sequence and structure properties. For most RBPs, however, only a primary sequence motif has been determined, while the structure of the binding sites is uncharacterized. Results: We developed SSMART, an RNA motif finder that simultaneously models the primary sequence and the structural properties of the RNA targets sites. The sequence-structure motifs are represented as consensus strings over a degenerate alphabet, extending the IUPAC codes for nucleotides to account for secondary structure preferences. Evaluation on synthetic data showed that SSMART is able to recover both sequence and structure motifs implanted into 3'UTR-like sequences, for various degrees of structured/unstructured binding sites. In addition, we successfully used SSMART on high-throughput in vivo and in vitro data, showing that we not only recover the known sequence motif, but also gain insight into the structural preferences of the RBP. Availability and implementation: SSMART is freely available at https://ohlerlab.mdc-berlin.de/software/SSMART_137/. Supplementary information: Supplementary data are available at Bioinformatics online.

BACKGROUND: RNA editing is a co-transcriptional modification that increases the molecular diversity, alters secondary structure and protein coding sequences by changing the sequence of transcripts. The most common RNA editing modification is the single base substitution (A→I) that is catalyzed by the members of the Adenosine deaminases that act on RNA (ADAR) family. Typically, editing sites are identified as RNA-DNA-differences (RDDs) in a comparison of genome and transcriptome data from next-generation sequencing experiments. However, a method for robust detection of site-specific editing events from replicate RNA-seq data has not been published so far. Even more surprising, condition-specific editing events, which would show up as differences in RNA-RNA comparisons (RRDs) and depend on particular cellular states, are rarely discussed in the literature. RESULTS: We present JACUSA, a versatile one-stop solution to detect single nucleotide variant positions from comparing RNA-DNA and/or RNA-RNA sequencing samples. The performance of JACUSA has been carefully evaluated and compared to other variant callers in an in silico benchmark. JACUSA outperforms other algorithms in terms of the F measure, which combines precision and recall, in all benchmark scenarios. This performance margin is highest for the RNA-RNA comparison scenario. We further validated JACUSA's performance by testing its ability to detect A→I events using sequencing data from a human cell culture experiment and publicly available RNA-seq data from Drosophila melanogaster heads. To this end, we performed whole genome and RNA sequencing of HEK-293 cells on samples with lowered activity of candidate RNA editing enzymes. JACUSA has a higher recall and comparable precision for detecting true editing sites in RDD comparisons of HEK-293 data. Intriguingly, JACUSA captures most A→I events from RRD comparisons of RNA sequencing data derived from Drosophila and HEK-293 data sets. CONCLUSION: Our software JACUSA detects single nucleotide variants by comparing data from next-generation sequencing experiments (RNA-DNA or RNA-RNA). In practice, JACUSA shows higher recall and comparable precision in detecting A→I sites from RNA-DNA comparisons, while showing higher precision and recall in RNA-RNA comparisons.

MOTIVATION: Although peak finding in next-generation sequencing (NGS) datasets has been addressed extensively, there is no consensus on how to analyze and process biological replicates. Furthermore, most peak finders do not focus on accurate determination of enrichment site widths and are not widely applicable to different types of datasets. RESULTS: We developed JAMM (Joint Analysis of NGS replicates via Mixture Model clustering): a peak finder that can integrate information from biological replicates, determine enrichment site widths accurately and resolve neighboring narrow peaks. JAMM is a universal peak finder that is applicable to different types of datasets. We show that JAMM is among the best performing peak finders in terms of site detection accuracy and in terms of accurate determination of enrichment sites widths. In addition, JAMM's replicate integration improves peak spatial resolution, sorting and peak finding accuracy. AVAILABILITY AND IMPLEMENTATION: JAMM is available for free and can run on Linux machines through the command line: http://code.google.com/p/jamm-peak-finder.

MOTIVATION: High-throughput sequencing of RNA in vivo facilitates many applications, not the least of which is the cataloging of variant splice isoforms of protein-coding messenger RNAs. Although many solutions have been proposed for reconstructing putative isoforms from deep sequencing data, these generally take as their substrate the collective alignment structure of RNA-seq reads and ignore the biological signals present in the actual nucleotide sequence. The majority of these solutions are graph-theoretic, relying on a splice graph representing the splicing patterns and exon expression levels indicated by the spliced-alignment process. RESULTS: We show how to augment splice graphs with additional information reflecting the biology of transcription, splicing and translation, to produce what we call an ORF (open reading frame) graph. We then show how ORF graphs can be used to produce isoform predictions with higher accuracy than current state-of-the-art approaches. AVAILABILITY AND IMPLEMENTATION: RSVP is available as C++ source code under an open-source licence: http://ohlerlab.mdc-berlin.de/software/RSVP/.

MOTIVATION: Pre-mRNA cleavage and polyadenylation are essential steps for 3'-end maturation and subsequent stability and degradation of mRNAs. This process is highly controlled by cis-regulatory elements surrounding the cleavage/polyadenylation sites (polyA sites), which are frequently constrained by sequence content and position. More than 50% of human transcripts have multiple functional polyA sites, and the specific use of alternative polyA sites (APA) results in isoforms with variable 3'-untranslated regions, thus potentially affecting gene regulation. Elucidating the regulatory mechanisms underlying differential polyA preferences in multiple cell types has been hindered both by the lack of suitable data on the precise location of cleavage sites, as well as of appropriate tests for determining APAs with significant differences across multiple libraries. RESULTS: We applied a tailored paired-end RNA-seq protocol to specifically probe the position of polyA sites in three human adult tissue types. We specified a linear-effects regression model to identify tissue-specific biases indicating regulated APA; the significance of differences between tissue types was assessed by an appropriately designed permutation test. This combination allowed to identify highly specific subsets of APA events in the individual tissue types. Predictive models successfully classified constitutive polyA sites from a biologically relevant background (auROC = 99.6%), as well as tissue-specific regulated sets from each other. We found that the main cis-regulatory elements described for polyadenylation are a strong, and highly informative, hallmark for constitutive sites only. Tissue-specific regulated sites were found to contain other regulatory motifs, with the canonical polyadenylation signal being nearly absent at brain-specific polyA sites. Together, our results contribute to the understanding of the diversity of post-transcriptional gene regulation. AVAILABILITY: Raw data are deposited on SRA, accession numbers: brain SRX208132, kidney SRX208087 and liver SRX208134. Processed datasets as well as model code are published on our website: http://www.genome.duke.edu/labs/ohler/research/UTR/. CONTACT: [email protected].

MOTIVATION: Computational approaches for the annotation of phenotypes from image data have shown promising results across many applications, and provide rich and valuable information for studying gene function and interactions. While data are often available both at high spatial resolution and across multiple time points, phenotypes are frequently annotated independently, for individual time points only. In particular, for the analysis of developmental gene expression patterns, it is biologically sensible when images across multiple time points are jointly accounted for, such that spatial and temporal dependencies are captured simultaneously. METHODS: We describe a discriminative undirected graphical model to label gene-expression time-series image data, with an efficient training and decoding method based on the junction tree algorithm. The approach is based on an effective feature selection technique, consisting of a non-parametric sparse Bayesian factor analysis model. The result is a flexible framework, which can handle large-scale data with noisy incomplete samples, i.e. it can tolerate data missing from individual time points. RESULTS: Using the annotation of gene expression patterns across stages of Drosophila embryonic development as an example, we demonstrate that our method achieves superior accuracy, gained by jointly annotating phenotype sequences, when compared with previous models that annotate each stage in isolation. The experimental results on missing data indicate that our joint learning method successfully annotates genes for which no expression data are available for one or more stages.

MiRNA are about 22nt long small noncoding RNAs that post transcriptionally regulate gene expression in animals, plants and protozoa. Confident identification of MiRNA-Target Interactions (MTI) is vital to understand their function. Currently, several integrated programs and databases are available for animal miRNAs, the mechanisms of which are significantly different from plant miRNAs. Here we present imiRTP, an integrated MTI prediction and analysis toolkit for Arabidopsis thaliana. It features two important functions: (i) combination of several effective plant miRNA target prediction methods provides a sufficiently large MTI candidate set, and (ii) different filters allow for an efficient selection of potential targets. The modularity of imiRTP enables the prediction of high quality targets on genome-wide scale.

Advances in reporters for gene expression have made it possible to document and quantify expression patterns in 2D-4D. In contrast to microarrays, which provide data for many genes but averaged and/or at low resolution, images reveal the high spatial dynamics of gene expression. Developing computational methods to compare, annotate, and model gene expression based on images is imperative, considering that available data are rapidly increasing. We have developed a sparse Bayesian factor analysis model in which the observed expression diversity of among a large set of high-dimensional images is modeled by a small number of hidden common factors. We apply this approach on embryonic expression patterns from a Drosophila RNA in situ image database, and show that the automatically inferred factors provide for a meaningful decomposition and represent common co-regulation or biological functions. The low-dimensional set of factor mixing weights is further used as features by a classifier to annotate expression patterns with functional categories. On human-curated annotations, our sparse approach reaches similar or better classification of expression patterns at different developmental stages, when compared to other automatic image annotation methods using thousands of hard-to-interpret features. Our study therefore outlines a general framework for large microscopy data sets, in which both the generative model itself, as well as its application for analysis tasks such as automated annotation, can provide insight into biological questions.

MOTIVATION: Evolutionarily conserved non-coding genomic sequences represent a potentially rich source for the discovery of gene regulatory region such as transcriptional enhancers. However, detecting orthologous enhancers using alignment-based methods in higher eukaryotic genomes is particularly challenging, as regulatory regions can undergo considerable sequence changes while maintaining their functionality. RESULTS: We have developed an alignment-free method which identifies conserved enhancers in multiple diverged species. Our method is based on similarity metrics between two sequences based on the co-occurrence of sequence patterns regardless of their order and orientation, thus tolerating sequence changes observed in non-coding evolution. We show that our method is highly successful in detecting orthologous enhancers in distantly related species without requiring additional information such as knowledge about transcription factors involved, or predicted binding sites. By estimating the significance of similarity scores, we are able to discriminate experimentally validated functional enhancers from seemingly equally conserved candidates without function. We demonstrate the effectiveness of this approach on a wide range of enhancers in Drosophila, and also present encouraging results to detect conserved functional regions across large evolutionary distances. Our work provides encouraging steps on the way to ab initio unbiased enhancer prediction to complement ongoing experimental efforts. AVAILABILITY: The software, data and the results used in this article are available at http://www.genome.duke.edu/labs/ohler/research/transcription/fly_enhancer/.

MOTIVATION: Recent advancements in high-throughput imaging have created new large datasets with tens of thousands of gene expression images. Methods for capturing these spatial and/or temporal expression patterns include in situ hybridization or fluorescent reporter constructs or tags, and results are still frequently assessed by subjective qualitative comparisons. In order to deal with available large datasets, fully automated analysis methods must be developed to properly normalize and model spatial expression patterns. RESULTS: We have developed image segmentation and registration methods to identify and extract spatial gene expression patterns from RNA in situ hybridization experiments of Drosophila embryos. These methods allow us to normalize and extract expression information for 78,621 images from 3724 genes across six time stages. The similarity between gene expression patterns is computed using four scoring metrics: mean squared error, Haar wavelet distance, mutual information and spatial mutual information (SMI). We additionally propose a strategy to calculate the significance of the similarity between two expression images, by generating surrogate datasets with similar spatial expression patterns using a Monte Carlo swap sampler. On data from an early development time stage, we show that SMI provides the most biologically relevant metric of comparison, and that our significance testing generalizes metrics to achieve similar performance. We exemplify the application of spatial metrics on the well-known Drosophila segmentation network. AVAILABILITY: A Java webstart application to register and compare patterns, as well as all source code, are available from: http://tools.genome.duke.edu/generegulation/image_analysis/insitu CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The computational detection of regulatory elements in DNA is a difficult but important problem impacting our progress in understanding the complex nature of eukaryotic gene regulation. Attempts to utilize cross-species conservation for this task have been hampered both by evolutionary changes of functional sites and poor performance of general-purpose alignment programs when applied to non-coding sequence. We describe a new and flexible framework for modeling binding site evolution in multiple related genomes, based on phylogenetic pair hidden Markov models which explicitly model the gain and loss of binding sites along a phylogeny. We demonstrate the value of this framework for both the alignment of regulatory regions and the inference of precise binding-site locations within those regions. As the underlying formalism is a stochastic, generative model, it can also be used to simulate the evolution of regulatory elements. Our implementation is scalable in terms of numbers of species and sequence lengths and can produce alignments and binding-site predictions with accuracy rivaling or exceeding current systems that specialize in only alignment or only binding-site prediction. We demonstrate the validity and power of various model components on extensive simulations of realistic sequence data and apply a specific model to study Drosophila enhancers in as many as ten related genomes and in the presence of gain and loss of binding sites. Different models and modeling assumptions can be easily specified, thus providing an invaluable tool for the exploration of biological hypotheses that can drive improvements in our understanding of the mechanisms and evolution of gene regulation.

MOTIVATION: The modeling of conservation patterns in genomic DNA has become increasingly popular for a number of bioinformatic applications. While several systems developed to date incorporate context-dependence in their substitution models, the impact on computational complexity and generalization ability of the resulting higher order models invites the question of whether simpler approaches to context modeling might permit appreciable reductions in model complexity and computational cost, without sacrificing prediction accuracy. RESULTS: We formulate several alternative methods for context modeling based on windowed Bayesian networks, and compare their effects on both accuracy and computational complexity for the task of discriminating functionally distinct segments in vertebrate DNA. Our results show that substantial reductions in the complexity of both the model and the associated inference algorithm can be achieved without reducing predictive accuracy.

DOAJ is a unique and extensive index of diverse open access journals from around the world, driven by a growing community, committed to ensuring quality content is freely available online for everyone.

BACKGROUND: Identifying functional elements, such as transcriptional factor binding sites, is a fundamental step in reconstructing gene regulatory networks and remains a challenging issue, largely due to limited availability of training samples. RESULTS: We introduce a novel and flexible model, the Optimized Mixture Markov model (OMiMa), and related methods to allow adjustment of model complexity for different motifs. In comparison with other leading methods, OMiMa can incorporate more than the NNSplice's pairwise dependencies; OMiMa avoids model over-fitting better than the Permuted Variable Length Markov Model (PVLMM); and OMiMa requires smaller training samples than the Maximum Entropy Model (MEM). Testing on both simulated and actual data (regulatory cis-elements and splice sites), we found OMiMa's performance superior to the other leading methods in terms of prediction accuracy, required size of training data or computational time. Our OMiMa system, to our knowledge, is the only motif finding tool that incorporates automatic selection of the best model. OMiMa is freely available at 1. CONCLUSION: Our optimized mixture of Markov models represents an alternative to the existing methods for modeling dependent structures within a biological motif. Our model is conceptually simple and effective, and can improve prediction accuracy and/or computational speed over other leading methods.

The split structure of most mammalian protein-coding genes allows for the potential to produce multiple different mRNA and protein isoforms from a single gene locus through the process of alternative splicing (AS). We propose a computational approach called UNCOVER based on a pair hidden Markov model to discover conserved coding exonic sequences subject to AS that have so far gone undetected. Applying UNCOVER to orthologous introns of known human and mouse genes predicts skipped exons or retained introns present in both species, while discriminating them from conserved noncoding sequences. The accuracy of the model is evaluated on a curated set of genes with known conserved AS events. The prediction of skipped exons in the approximately 1% of the human genome represented by the ENCODE regions leads to more than 50 new exon candidates. Five novel predicted AS exons were validated by RT-PCR and sequencing analysis of 15 introns with strong UNCOVER predictions and lacking EST evidence. These results imply that a considerable number of conserved exonic sequences and associated isoforms are still completely missing from the current annotation of known genes. UNCOVER also identifies a small number of candidates for conserved intron retention.

In our paper we present two new approaches for language identification. Both of them are based on the use of so-called multigrams, an information theoretic based observation representation. In the first approach we use multigram models for phonotactic modeling of phoneme or codebook sequences. The multigram model can be used to segment the new observation into larger units (e.g. something like words) and calculates a probability for the best segmentation. In the second approach we build a fenon recognizer using the segments of the best segmentation of the training material as "words" inside the recognition vocabulary. On the OGI test corpus and on the NIST'95 evaluation corpus we got significant improvements with this second approach in comparison to the unsupervised codebook approach when discriminating between English and German utterances.

We show how discriminative training methods, namely the Maximum Mutual Information and Maximum Discrimination approach, can be adopted for the training of N-gram language models used as classifiers working on symbol strings. By estimating the model parameters according to a discriminative objective function instead of Maximum Likelihood, the emphasis is not put on the exact modeling of each class, but on the right classification of the samples. The methods are shown to be suited for a variety of applications, such as the recognition of regulatory DNA sequences and language identification. Using phonotactic information, we achieve an error reduction of 10.7% (phoneme sequences) or 41.9% (codebook classes) with respect to the standard ML estimation on a corpus of English and German sentences. 1.

MOTIVATION: We describe a new content-based approach for the detection of promoter regions of eukaryotic protein encoding genes. Our system is based on three interpolated Markov chains (IMCs) of different order which are trained on coding, non-coding and promoter sequences. It was recently shown that the interpolation of Markov chains leads to stable parameters and improves on the results in microbial gene finding (Salzberg et al., Nucleic Acids Res., 26, 544-548, 1998). Here, we present new methods for an automated estimation of optimal interpolation parameters and show how the IMCs can be applied to detect promoters in contiguous DNA sequences. Our interpolation approach can also be employed to obtain a reliable scoring function for human coding DNA regions, and the trained models can easily be incorporated in the general framework for gene recognition systems. RESULTS: A 5-fold cross-validation evaluation of our IMC approach on a representative sequence set yielded a mean correlation coefficient of 0.84 (promoter versus coding sequences) and 0.53 (promoter versus non-coding sequences). Applied to the task of eukaryotic promoter region identification in genomic DNA sequences, our classifier identifies 50% of the promoter regions in the sequences used in the most recent review and comparison by Fickett and Hatzigeorgiou ( Genome Res., 7, 861-878, 1997), while having a false-positive rate of 1/849 bp.