Skip to content

Releases: WGLab/NanoCaller

v2.1.1

19 Feb 18:28
Compare
Choose a tag to compare

Fixed a bug that was crashing SNP calling threads trying to convert empty array into int array.

v2.1.0

15 Feb 04:28
Compare
Choose a tag to compare

Migrated NanoCaller models from tensorflow 1 to tensorflow 2, and updated python requirements as well.

v2.0.0 Major API Update

25 Jan 20:11
Compare
Choose a tag to compare
  1. Major changes in API to simplify run commands, with PATH variable updated automatically upon environment creation.
  2. Create summary of variant calling results using RTGtools upon completion
  3. A new parameter --impute_indel_phase allows indel calling when phasing information cannot be provided
  4. Fixed an indel calling bug that affected indel prediction near boundaries of regions specified

v1.0.1

15 Sep 18:20
Compare
Choose a tag to compare

Fixed issues with indel calling near the end of reference contigs.

v1.0.0

10 Aug 17:08
bc158db
Compare
Choose a tag to compare

First post-production release with Zenodo DOI citation.

DOI

v0.4.1

03 Aug 20:05
Compare
Choose a tag to compare

Reverted changes from commit: 2546959 which used python's map function instead of multiprocessing pool map and caused the code to run slower.

v0.4.0

02 Jun 20:16
Compare
Choose a tag to compare

Major update in NanoCaller SNP and indel calling routines. Several new models trained on ONT reads basecalled with Guppy4.2.2 and Bonito are added, as well as a new model for R10.3 reads. SNP models trained on longer CCS reads are also added. Indel calling is improved to increase sensitivity of indels in low complexity regions.

v0.3.3

11 Mar 16:43
20061e3
Compare
Choose a tag to compare

Enabled error logging for parallel jobs run during whole genome variant calling, with details of failed jobs.

v0.3.2

15 Jan 22:01
Compare
Choose a tag to compare

Fixed some bugs with chromosome notation.

v0.3.1 Fixed issues with excessive processes created by python mutliprocessing

15 Dec 18:01
Compare
Choose a tag to compare

Disabled python multiprocessing pooling when single CPU is used. This creates half the number of processes when using NanoCaller_WGS.py for whole genome variant calling, leading to less overhead.