Tamburrini, A. | bioRxiv Science

bioRxiv2022

RNABSdb and 3plex enable deep computational investigation of triplex forming lncRNAs

1.Long non-coding RNAs (lncRNAs) regulate gene expression through different molecular mechanisms, including DNA binding. We curated the first database of RNA Binding Sites (RNABSdb) by harmonising publicly available raw-data of RNA-DNA binding experiments. This resource is crucial to enable systematic studies on transcriptional regulation driven by lncRNAs. Focusing on high quality experiments, we find that the number of binding sites for each lncRNAs varies from hundreds to tens of thousands. Despite being poorly characterised, the formation of RNA:DNA:DNA triple helices (TPXs) is one of the molecular mechanisms that allows lncRNAs to bind the genome and regulate gene expression. We developed 3plex, a software able to predict TPXs in silico. We show that 3plex outperforms previous existing approaches leveraging the data collected in RNABSdb for lncRNAs known to form functional TPXs. Moreover this analysis shows that TPXs tend to be shorter and more degenerated than previously expected. Finally, we applied 3plex to all the lncRNAs collected in RNABSdb and we show that the majority of them could directly bind the genome by TPXs formation. Data and software are available at https://molinerislab.github.io/RNABSdb/ and https://github.com/molinerisLab/3plex.

bioinformatics↗

bioRxiv2019

Janggu: Deep Learning for Genomics

MotivationIn recent years, numerous applications have demonstrated the potential of deep learning for an improved understanding of biological processes. However, most deep learning tools developed so far are designed to address a specific question on a fixed dataset and/or by a fixed model architecture. Adapting these models to integrate new datasets or to address different hypotheses can lead to considerable software engineering effort. To address this aspect we have built Janggu, a python library that facilitates deep learning for genomics applications.\n\nResultsJanggu aims to ease data acquisition and model evaluation in multiple ways. Among its key features are special dataset objects, which form a unified and flexible data acquisition and pre-processing framework for genomics data that enables streamlining of future research applications through reusable components. Through a numpy-like interface, the dataset objects are directly compatible with popular deep learning libraries, including keras. Furthermore, Janggu offers the possibility to visualize predictions as genomic tracks or by exporting them to the BIGWIG format. We illustrate the functionality of Janggu on several deep learning genomics applications. First, we evaluate different model topologies for the task of predicting binding sites for the transcription factor JunD. Second, we demonstrate the framework on published models for predicting chromatin effects. Third, we show that promoter usage measured by CAGE can be predicted using DNase hyper-sensitivity, histone modifications and DNA sequence features. We improve the performance of these models due to a novel feature in Janggu that allows us to include high-order sequence features. We believe that Janggu will help to significantly reduce repetitive programming overhead for deep learning applications in genomics, while at the same time enabling computational biologists to assess biological hypotheses more rapidly.\n\nAvailabilityJanggu is freely available under a GPL-v3 license on https://github.com/BIMSBbioinfo/janggu or via https://pypi.org/project/janggu

bioinformatics↗