A new parallel pipeline for DNA methylation analysis of long reads datasets

HPG-DHunter: an ultrafast, friendly tool for DMR detection and visualization

BACKGROUND

Software tools for analyzing DNA methylation do not provide graphical results which can be easily identified, but huge text files containing the alignment of the samples and their methylation status at a resolution of base pairs. There have been proposed different tools and methods for finding Differentially Methylated Regions (DMRs) among different samples, but the execution time required by these tools is large, and the visualization of their results is far from being interactive. Additionally, these methods show more accurate results when identifying simulated DM regions that are long and have small within-group variation, but they have low concordance when used with real datasets, probably due to the different approaches they use for DMR identification. Thus, a tool which automatically detects DMRs among different samples and interactively visualizes DMRs at different scales (from a bunch to ten of millions of DNA locations) can be the key for shortening the DNA methylation analysis process in many studies.

RESULTS

In this paper, we propose a software tool based on the wavelet transform. This mathematical tool allows the fast automatic DMR detection by simple comparison of different signals at different resolution levels. Also, it allows an interactive visualization of the DMRs found at different resolution levels. The tool is publicly available at https://grev-uv.github.io/ , and it is part of a complete suite of tools which allow to carry out the complete process of DNA alignment and methylation analysis, creation of methylation maps of the whole genome, and the detection and visualization of DMRs between different samples.

CONCLUSIONS

The validation of the developed software tool shows similar concordance with other well-known and extended tools when used with real and synthetic data. The batch mode of the tool is capable of automatically detecting the existing DMRs for half (twelve) of the human chromosomes between two sets of six samples (whose.csv files after the alignment and mapping procedures have an aggregated size of 108 Gigabytes) in around three hours and a half. When compared to other well-known tools, HPG-DHunter only requires around 15% of the execution time required by other tools for detecting the DMRs.

DNA methylation studies of depression with onset in the peripartum: A critical systematic review

OBJECTIVE

Major depression with peripartum onset (MDP) has been associated with multiple adverse offspring health outcomes. The biological mechanisms underlying this relationship remain unclear, but DNA methylation (DNAm) represents a plausible mechanism for mediating MDP exposures and changes in offspring development, behavior, and health. Advances in DNAm research necessitate reevaluating the MDP-DNAm literature to determine how well past studies conform with current best practices.

METHOD

Five databases were searched to identify studies of prenatal-onset MDP and DNAm. Quality scores were assigned to each article independently by two raters using a novel scale specific for MDP-DNAm research.

RESULTS

Nineteen studies met inclusion criteria. Quality scores ranged from 10 to 17 out of 24 points (M = 12.8; SD = 1.9), with higher scores indicating increased study rigor. Poor covariate reporting was the most significant contributor to lower scores.

CONCLUSION

No longitudinal MDP-DNAm studies exist. Earlier MDP-DNAm studies should be interpreted with caution, and future research must commit to sharing methodology and data to facilitate cross-study comparisons and maximize dataset utility.