Back

TSSpredator-Web: A web-application for transcription start site prediction and exploration

Witte Paz, M.; Herbig, A.; Nieselt, K.

2025-06-01 bioinformatics
10.1101/2025.05.29.656934 bioRxiv
Show abstract

BackgroundWith the rapid development of high-throughput RNA-seq technologies, the transcriptome of prokaryotes can now be studied in unprecedented detail. Transcription start site (TSS) identification provides critical insights into transcriptional regulation, yet current command-line tools for the prediction of TSS remain challenging with respect to their usability and lack of integrated exploration features. ResultsWe introduce TSSpredator-Web an interactive web application that enhances the usability of the established yet unpublished tool, TSSpredator. TSSpredator-Web facilitates TSS prediction from non-enriched and enriched RNA-seq data, classifies TSS relative to annotated genes, and allows users to explore results through dynamic visualizations and interactive tables. For the visualizations we provide an UpSet plot summarizing TSS distribution across experiments or classes, and a genome viewer that integrates transcriptomic and genomic data that contextualizes the insights of the TSS predictions. To illustrate the usage of TSSpredator-Web, we provide a use case with Cappable-seq data from Escherichia coli. TSSpredator-Web is available at the TueVis visualization web-server at https://tsspredator-tuevis.cs.uni-tuebingen.de/. ConclusionsBy combining user-friendly accessibility with interactive data exploration, TSSpredator-Web significantly facilitates genome-wide TSS analysis and interpretation in prokaryotes, empowering a broader range of researchers to generate biological insights from transcriptomic data.

Published in PLOS ONE (predicted rank #9) · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.