Understanding gene regulation is a key challenge in today's biology. The new technologies of protein-binding microarrays (PBMs) and high-throughput SELEX (HT-SELEX) allow measurement of the binding intensities of one transcription factor (TF) to numerous synthetic double-stranded DNA sequences in a single experiment. Recently, Jolma et al. reported the results of 547 HT-SELEX experiments covering human and mouse TFs. Because 162 of these TFs were also covered by PBM technology, for the first time, a large-scale comparison between implementations of these two in vitro technologies is possible. Here we assessed the similarities and differences between binding models, represented as position weight matrices, inferred from PBM and HT-SELEX, and also measured how well these models predict in vivo binding. Our results show that HT-SELEX- and PBM-derived models agree for most TFs. For some TFs, the HT-SELEX-derived models are longer versions of the PBM-derived models, whereas for other TFs, the HT-SELEX models match the secondary PBM-derived models. Remarkably, PBM-based 8-mer ranking is more accurate than that of HT-SELEX, but models derived from HT-SELEX predict in vivo binding better. In addition, we reveal several biases in HT-SELEX data including nucleotide frequency bias, enrichment of C-rich k-mers and oligos and underrepresentation of palindromes.

Transcription factors (TFs) achieve DNA‐binding specificity through contacts with functional groups of bases (base readout) and readout of structural properties of the double helix (shape readout). Currently, it remains unclear whether DNA shape readout is utilized by only a few selected TF families, or whether this mechanism is used extensively by most TF families. We resequenced data from previously published HT‐SELEX experiments, the most extensive mammalian TF–DNA binding data available to date. Using these data, we demonstrated the contributions of DNA shape readout across diverse TF families and its importance in core motif‐flanking regions. Statistical machine‐learning models combined with feature‐selection techniques helped to reveal the nucleotide position‐dependent DNA shape readout in TF‐binding sites and the TF family‐specific position dependence. Based on these results, we proposed novel DNA shape logos to visualize the DNA shape preferences of TFs. Overall, this work suggests a way of obtaining mechanistic insights into TF–DNA binding without relying on experimentally solved all‐atom structures.

Significance: Transcription factors (TFs) are key proteins that bind DNA targets to coordinate gene expression in cells. Understanding how TFs recognize their DNA targets is essential for predicting how variations in regulatory sequence disrupt transcription to cause disease. Here, we develop a high-throughput assay and analysis pipeline capable of measuring binding energies for over one million sequences with high resolution and apply it toward understanding how nucleotides flanking DNA targets affect binding energies for two model yeast TFs. Through systematic comparisons between models trained on these data, we establish that considering dinucleotide (DN) interactions is sufficient to accurately predict binding and further show that sites used by TFs in vivo are both energetically and mutationally distant from the highest affinity sequence.

The minimizers scheme is a method for selecting k-mers from sequences. It is used in many bioinformatics software tools to bin comparable sequences or to sample a sequence in a deterministic fashion at approximately regular intervals, in order to reduce memory consumption and processing time. Although very useful, the minimizers selection procedure has undesirable behaviors (e.g. too many k-mers are selected when processing certain sequences). Some of these problems were already known to the authors of the minimizers technique, and the natural lexicographic ordering of k-mers used by minimizers was recognized as their origin. Many software tools using minimizers employ ad hoc variations of the lexicographic order to alleviate those issues. We provide an in-depth analysis of the effect of k-mer ordering on the performance of the minimizers technique. By using small universal hitting sets (a recently defined concept), we show how to significantly improve the performance of minimizers and avoid some of its worse behaviors. Based on these results, we encourage bioinformatics software developers to use an ordering based on a universal hitting set or, if not possible, a randomized ordering, rather than the lexicographic order. This analysis also settles negatively a conjecture (by Schleimer et al.) on the expected density of minimizers in a random sequence. The software used for this analysis is available on GitHub: https://github.com/gmarcais/minimizers.git.

Motivation: Protein–RNA interactions, which play vital roles in many processes, are mediated through both RNA sequence and structure. CLIP-based methods, which measure protein–RNA binding in vivo, suffer from experimental noise and systematic biases, whereas in vitro experiments capture a clearer signal of protein RNA-binding. Among them, RNAcompete provides binding affinities of a specific protein to more than 240 000 unstructured RNA probes in one experiment. The computational challenge is to infer RNA structure- and sequence-based binding models from these data. The state-of-the-art in sequence models, Deepbind, does not model structural preferences. RNAcontext models both sequence and structure preferences, but is outperformed by GraphProt. Unfortunately, GraphProt cannot detect structural preferences from RNAcompete data due to the unstructured nature of the data, as noted by its developers, nor can it be tractably run on the full RNACompete dataset. Results: We develop RCK, an efficient, scalable algorithm that infers both sequence and structure preferences based on a new k-mer based model. Remarkably, even though RNAcompete data is designed to be unstructured, RCK can still learn structural preferences from it. RCK significantly outperforms both RNAcontext and Deepbind in in vitro binding prediction for 244 RNAcompete experiments. Moreover, RCK is also faster and uses less memory, which enables scalability. While currently on par with existing methods in in vivo binding prediction on a small scale test, we demonstrate that RCK will increasingly benefit from experimentally measured RNA structure profiles as compared to computationally predicted ones. By running RCK on the entire RNAcompete dataset, we generate and provide as a resource a set of protein–RNA structure-based models on an unprecedented scale. Availability and Implementation: Software and models are freely available at http://rck.csail.mit.edu/ Contact:bab@mit.edu Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: The complexes formed by binding of proteins to RNAs play key roles in many biological processes, such as splicing, gene expression regulation, translation and viral replication. Understanding protein-RNA binding may thus provide important insights to the functionality and dynamics of many cellular processes. This has sparked substantial interest in exploring protein-RNA binding experimentally, and predicting it computationally. The key computational challenge is to efficiently and accurately infer protein-RNA binding models that will enable prediction of novel protein-RNA interactions to additional transcripts of interest. Results: We developed DLPRB (Deep Learning for Protein-RNA Binding), a new deep neural network (DNN) approach for learning intrinsic protein-RNA binding preferences and predicting novel interactions. We present two different network architectures: a convolutional neural network (CNN), and a recurrent neural network (RNN). The novelty of our network hinges upon two key aspects: (i) the joint analysis of both RNA sequence and structure, which is represented as a probability vector of different RNA structural contexts; (ii) novel features in the architecture of the networks, such as the application of RNNs to RNA-binding prediction, and the combination of hundreds of variable-length filters in the CNN. Our results in inferring accurate RNA-binding models from high-throughput in vitro data exhibit substantial improvements, compared to all previous approaches for protein-RNA binding prediction (both DNN and non-DNN based). A more modest, yet statistically significant, improvement is achieved for in vivo binding prediction. When incorporating experimentally-measured RNA structure, compared to predicted one, the improvement on in vivo data increases. By visualizing the binding specificities, we can gain biological insights underlying the mechanism of protein RNA-binding. Availability and implementation: The source code is publicly available at https://github.com/ilanbb/dlprb. Supplementary information: Supplementary data are available at Bioinformatics online.