Release of Stacks v2.66
Catchen, Julian
Hi All,
I am happy to release Stacks version 2.66 today. The main change is I found the error causing segfaults when running process_radtags in multi-threaded mode. In short, when processing paired-end reads, asking for discarded reads to be retained (-D) could cause a crash as having the multiple threads writing to the single discards file caused conflicts. Now the program will write one discard file per output thread, which is more efficient than locking the file to be written by a single thread at a time.
I also improved some of the supplementary Python scripts and include the stacks-private-alleles script for the first time. This script allows one to analyze private alleles between two ‘parental’ populations and one or more ‘hybrid’ populations (defining ‘parent’ to mean a population with private alleles and ‘hybrid’ to mean a different population that may contain a mix of ‘parent’ private alleles). The script will allow you to plot these private alleles in two ways along the chromosomes of a genome; in the hybrids it will plot 1) private alleles from one ‘parent’ on the background of the other (and vice versa) and 2) it will do kernel smoothing of private allele frequencies along the chromosomes.
I also improved the stacks-dist-extract script, so if you haven’t tried it yet, I recommend it for making the log data more accessible and easier to understand.
I have added pages to the website (e.g. https://catchenlab.life.illinois.edu/stacks/comp/stacks_dist_extract.php) describing these scripts and will (eventually) post a tutorial on private alleles (here are some bluesky posts on how it works: https://bsky.app/profile/jcatchen.bsky.social/post/3kcjnlfgs3c2o).
Stacks 2.66 - December 5, 2023
------------------------------
Feature: Rewrote stacks-dist-extract in Python including new support for partial section names, streaming capability, and other improvements.
Feature: Included new stacks-private-alleles script that will extract private allele data from the populations program outputs and output useful summaries and prepare it for plotting.
Bugfix: In clone_filter, users sometimes specified a single oligo sequence on the paired read, but the length of that oligo with --oligo-len-2 instead of --oligo-len-1. Added code to use oligo length from either parameter when a single sequence is specified.
Bugfix: private allele summary count in populations.log could be incorrect, values in populations.sumstats.tsv were not affected.
Bugfix: when running in parallel with paired-end reads and retaining discarded reads, process_radtags could segfault. Corrected threads writing to discard files. Updated naming of discard output file.
Bugfix: corrected two small memory access errors in process_radtags.
Best,
julian
Ps – I have two Blue Sky invites if anyone wants one just let me know.
