v2.2.0

Earlier versions of guidescan enumerate --format sam ... assumed that the kmer chromosome and position can be ascertained directly from the incoming kmer file. While this is reasonable, a lot of times this information is not available beforehand, and blank/dummy values for chromosome/position are filled in only to satisfy guidescan enumerate.

In these cases, the output of guidescan enumerate --format csv ... is correct since it obtains the match positions directly from the offtargets (including at distance 0), but the output of guidescan enumerate --format sam ... is incorrect w.r.t. the reference name and reference position fields in the SAM file (because that information was unavailable to begin with).

This release fixes this issue by getting that information directly from the off-targets at distance 0.

Note that this means that multiple matches at distance 0 will end up producing multiple lines in the SAM file, with identical off-target hex information. This is already happening in the CSV file generation, so this also makes the behavior consistent.

v2.1.8

This release of guidescan fixes a minor bug when using guidescan enumerate with options --format csv --mode complete.

In case no matches and no offtargets of the input gRNA were found, the order of the last 4 output csv fields would have been incorrect. This is fixed now.

This release also generates csv files if gRNA sequences are not exactly 23 NUCs long, instead of crashing. This is to allow free-form search of gRNAs against an index. Specificity calculations in such cases should be interpreted carefully of course, since cutting efficiency/specificity are technically defined only for 20mers with NGG PAMs.

v2.1.7

This release of guidescan fixes two issues:

• In the case of 0 matches and 0 mismatches of a kmer, guidescan enumerate --format csv previously failed to produce any output line corresponding to the kmer. This is now fixed, and an output line is generated for the missing kmer.
• gRNAs with only a match with the NAG pam (and no matches with the NGG pam), would be reported with specificity > 1. In these cases, the match is now classified as an imperfect match, and a value of 1 is added to the CFD scores before taking the reciprocal, to allow all specificity values to be <= 1.

v2.1.6

This release fixes a bug in determining specificity values during guidescan enumerate. The incorrect specificity values would have shown up if --alt-pam flag was specified with guidescan enumerate. The code was initially looking at the PAM obtained from the kmer file, but now we look at the actual 3-nucleotide sequence at the end of the match. Of course, the specificity values are understood to make sense only in case of cas9.

v2.1.4

Generating correct src tar.gz for subsequent bioconda recipe integration

v2.1.3 Release

What's Changed

Bug fixes and default parameter improvements

• Updated the cmake min. version by @vineetbansal in #17
  Full Changelog: v2.1.2...v2.1.3

v2.1.2

Installing libcurl for Linux (#16)

* Added a workflow to check successful building of guidescan
* Installing libcurl on linux; Using guidescan --version as a test command
* Installing libcurl on release workflow

v2.0.0 Release

Major Changes

• We will now be versioning our software starting with version 2.0.0, this is the first release.

What's Changed

• Specificity formula fix by @dimitarch in #4
• Minor changes to the manual by @vineetbansal in #6
• Bug fix in cfd calculation by @vineetbansal in #8
• Guidescan versioning and release process by @vineetbansal in #10

New Contributors

• @dimitarch made their first contribution in #4
• @vineetbansal made their first contribution in #6

Full Changelog: https://github.com/schmidt73/guidescan-cli/commits/v2.0.0