Discovering sequence and structure landscapes in RNA interaction motifs

Abstract

RNA molecules are able to bind proteins, DNA and other small or long RNAs using information at primary, secondary or tertiary structure level. Recent techniques that use cross-linking and immunoprecipitation of RNAs can detect these interactions and, if followed by high-throughput sequencing, molecules can be analysed to find recurrent elements shared by interactors, such as sequence and/or structure motifs. Many tools are able to find sequence motifs from lists of target RNAs, while others focus on structure using different approaches to find specific interaction elements. In this work, we make a systematic analysis of RBP–RNA and RNA–RNA datasets to better characterize the interaction landscape with information about multi-motifs on the same RNAs. To achieve this goal, we updated our BEAM algorithm to combine both sequence and structure information to create pairs of patterns that model motifs of interaction. This algorithm was applied to several RNA binding proteins and ncRNAs interactors, confirming already known motifs and discovering new ones. This landscape analysis on interaction variability reflects the diversity of target recognition and underlines that often both primary and secondary structure are involved in molecular recognition.

ELAVL1 primarily couples mRNA stability with the 3′ UTRs of interferon-stimulated genes

“…For 27 different RBPs (including ELAVL1), the preferred 6-mer binding site overlapped with the top-ranked 6-mer site of at least one other RBP, despite having distinct RNA-binding domains (Dominguez et al, 2018). This observation underscores that many RBPs have similar and short binding motif preferences of low complexity (Adinolfi et al, 2019;Nussbacher and Yeo, 2018). Although these studies are valuable in discovering the primary and flanking sequence preferences of RBPs, these methods often analyze a single RBP in isolation and do not include competitor RBPs and miRNAs that will influence the RNA-substrate structure or the availability of a particular RNA binding site.…”

Big data analytics in bioinformatics

“…DeepCLIP produces motifs of varying sizes ranked by the average information content (Figure 2a, b). The top-ranking motifs of DeepCLIP for the analyzed proteins were visually remarkably similar to core binding sites as described in literature (67,68,72,(78)(79)(80)(81)(82) (Figure S6). The motifs depicted in Figure S6 are based on the top-1000 scoring sequences from the positive and negative input dataset and represent the two pseudo-PFMs with the highest mean information score among the five pseudo-PFMs produced.…”

“…In the following, all inputs to and outputs of presented neural network layers will be in vector form unless stated otherwise, as this might ease the reader's understanding of the math behind the computation of the layer outputs. kernels or filters (both are alternative terms for the weight matrices of convolutional nodes) [81][82][83]. Convolutional nodes can stride along presented input examples and thereby scan the data instances for important features.…”

Intragenic MicroRNAs Autoregulate Their Host Genes in Both Direct and Indirect Ways—A Cross-Species Analysis

“…This may explain why performance of bioinformatics studies based on nucleotide sequences cannot be sufficiently precise on its own. Additional parameters, such as three-dimensional structures and binding energy of specific miRNA motifs, need to be taken into account as they become available for a full understanding of miRNA::host gene autoregulation [84]. However, and in line with studies reporting that individual miRNAs target disease-relevant molecular pathways [60,64,66] our unbiased genome-wide study provides important evidence that network-wide targeting is not the result of random associations but more likely resembles a generic mechanism regulating entire molecular pathways [85].…”