Derbel, H. | bioRxiv Science

bioRxiv2025

DiffMethylTools: a toolbox of the detection, annotation and visualization of differential DNA methylation

DNA methylation is a fundamental epigenetic mechanism, and its significant changes (i.e., differential methylation) regulate gene expression, cell-type specification and disease progression without altering the underlying DNA sequence. Differential methylation was usually detected via existing statistical tools by comparing two groups of methyomes (i.e. whole-genome methylation profiles) and has wide applications of various downstream investigations for human disease studies. However, few toolboxes were available to efficiently streamline methylation investigation by integrating robust detection, annotation and visualization of differential methylation. Also, differential methylation detected via tools has poor reproducibility and no tools were tested on the increasing volume of long read methylomes. To address these issues, we introduced DiffMethylTools, an end-to-end solution to eliminate analytical and computational difficulties for differential methylation dissection. Comparison of detection performance on six datasets including three long-read methylomes demonstrated that DiffMethylTools achieved overall better performance of detecting differential methylation than existing tools like MethylKit, DSS, MethylSig, and bsseq. Besides, DiffMethylTools supported versatile input formats for seamless transition from upstream methylation detection tools, and offered diverse annotations and visualizations to facilitate downstream investigations. DiffMethylTools therefore offered a robust, interpretable, and user-friendly solution for differential methylation investigation, benefiting the dissection of methylations roles in human disease studies.

bioinformatics↗

bioRxiv2023

Accurate prediction of transcriptional activity of single missense variants in HIV Tat with deep learning

Tat is an essential gene for increasing the transcription of all HIV genes, and it affects HIV replication, HIV exit from latency, and AIDS progression. The Tat gene frequently mutates in vivo producing variants with diverse activities, contributing to HIV viral heterogeneity, as well as drug-resistant clones. Thus, identifying the transcriptional activities of Tat variants will help to better understand AIDS pathology and treatment. We recently reported the missense mutation landscape of all single amino acid Tat variants. In these experiments, a fraction of double missense alleles exhibited intragenic epistasis. It is too time-consuming and costly to determine a variants effect for all double mutant alleles with experiments. Therefore, we propose a combined GigaAssay/Deep learning approach. As a first step for determining activity landscapes for complex variants, we evaluated a deep learning framework using previously reported GigaAssay experiments to predict how transcription activity is affected by Tat variants with single missense substitutions. Our approach achieves a 0.94 Pearson correlation coefficient when comparing experimental to predicted activities. This hybrid approach should be extensible to more complex Tat alleles for better understanding the genetic control of HIV genome transcription.

bioinformatics↗