Skip to content

DNAnalyzer v3.5.0-beta.0

Latest
Compare
Choose a tag to compare
@VerisimilitudeX VerisimilitudeX released this 24 Dec 23:28
· 13 commits to main since this release
921f63c

Introduction

This release introduces full batch FASTA file support, quantitative insights into disease-associated gene expression sites, annotations for CFTR sequences, and parallelized codon mapping for even faster data processing. Additionally, it addresses several major compatibility bugs on Linux and uses Multi-Task Learning (MTL) to further improve performance by learning multiple machine learning tasks at the same time.

DNAnalyzer currently takes just 11 minutes to conduct analyses on a full human genome running on an M4 Max MacBook Pro, a 47% reduction on average from v3.0.0-beta.0. The end goal for DNAnalyzer is to be able to run on a mobile device in mere seconds.

Of course, this release would not have been possible without contributions from the computational biology research community, and we are grateful to the 43 returning and first-time contributors for their commits, discussion posts, and issues. Please continue reading for more updates and the full change log.

Updates to Citations

If you utilize this software for personal, institutional, or company purposes, please cite DNAnalyzer using:

Acharya, P. (2022). Democratizing ML-powered DNA analysis through efficient on-device analysis and interpretive tools. (Version 3.5.0-beta.0) [Computer software]. https://github.com/VerisimilitudeX/DNAnalyzer

What's Changed

Features

Fixes

Code Improvements

Documentation Updates

Styling and Formatting

  • Style: format code with Autopep8 and Google Java Format by @deepsource-autofix in #450 and other PRs.
  • [ImgBot] Optimize images by @imgbot in #396

Miscellaneous


New Contributors


Full Changelog: v3.0.0-beta.0...v3.5.0-beta.0