Original title: pydreg: a fast Python package for identifying active cis -regulatory elements from nascent transcription
Adam He, Charles Danko · September 2026
Active control switches in our DNA, called promoters and enhancers, make distinct patterns of newly formed RNA. Scientists can read these patterns to find active regulatory switches across the genome using a program called dREG. However, the original dREG relied on an old software setup that is now hard to install and maintain. To solve this problem, researchers built pydreg, a new version written in the Python programming language. It uses the exact same peak-detecting steps and trained models as the original, but runs on modern computing tools. Because of this rewrite, pydreg produces nearly identical results to dREG. It also runs 4.5 times faster, uses 5.4 times less peak memory, and runs much more easily on local computers.
You are trying to find all the light switches in a huge building by listening for the hum of lights turning on. The original method required a special toolkit that only worked on one type of computer and took hours. Here, the researchers rebuilt the same listening method using common tools that run on any laptop, finishing in minutes instead of hours.
Researchers studying how genes are turned on and off need to find the DNA switches—promoters and enhancers—that control gene activity. pydreg makes this search faster and easier to run on a standard laptop instead of a specialized server, so more labs can study gene regulation without expensive computing setups.
The abstract does not evaluate pydreg on new datasets or compare its accuracy beyond noting nearly identical peak calls. It does not address whether the pretrained models generalize to different cell types or sequencing protocols. The speed and memory improvements are measured against the original dREG implementation, not against other methods.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
pydreg: a fast Python package for identifying active cis -regulatory elements from nascent transcription — Adam He, Charles Danko — bioRxiv — doi:10.64898/2026.09.06.745329
Abstract licence: CC BY
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.