patch v2.1.2

[bkohrn]

v2.1.2: Bugfix:

• Re-add pre variant calling checkpoint to fix non-functionality of rerun mode 1.

Internal Changes:

• Instead of removing families containing Ns or mononucleotide repeats in UMIs after all consensus making happened, do it before any consensus making happens. The tagstats output removes unchanged at this time; the only difference will be in the SSCS output, which will not contain reads with UMIs that contain Ns or mononucleotide repeats (it did previously).
• Track number of N-containing UMIs in CM_stats file.
• Modify rerun to check for file existence (depends on OS module from Python standard library; may not work identically on all systems).

[bkohrn]

As far as I can tell, this seems to be working on WSL1. Before merging, it should be tested on WSL2, Linux, and Mac OS X (at a minimum).

[bkohrn]

I can probably manage a Mac OS X test.

[bkohrn]

Just realized I still need to update the README due to a slight change in the produced outputs.

[bkohrn]

Just realized: this UCM produces different output (something in the mono-nucleotide repeat detection isn't working properly). As stands (in v2.1.1), the cm_stats file looks like this:

Consensus Making Statistics:
Command: /home/kohrnb/bioinformatics/Duplex-Seq-Pipeline-v2.0.0/scripts/UnifiedConsensusMaker.py --input /dev/stdin --taglen 8 --spacerlen 1 --loclen 8 --write-sscs --prefix noProf.test_main.1868MCL --tagstats --cutoff 0.7 --Ncutoff 0.0142 --numCores 8 --minmem 3 --maxmem 200
Started at Friday, 25. June 2021 10:28AM
Finished at Friday, 25. June 2021 10:39AM
6055981 reads UMI processed
6055980 reads processed
1229764 families processed
    622052 unrepresented families
    566372 families with family size < 3
    1262 families (DCS pairs) filtered for UMIs with mononucleotide repeats
663392 SSCS made
    0 SSCS filtered for excessive Ns
216912 DCS made
    708996 DCS failed due to missing SSCS
    27228 DCS filtered for excessive Ns

The current dev version looks like this:

Consensus Making Statistics:
Command: /home/kohrnb/bioinformatics/dev/DS_main/Duplex-Seq-Pipeline/scripts/UnifiedConsensusMaker.py --input /dev/stdin --taglen 8 --spacerlen 1 --loclen 8 --write-sscs --prefix noProf.devTest_main.1868MCL --tagstats --cutoff 0.7 --Ncutoff 0.0142 --numCores 8 --minmem 3 --maxmem 200
Started at Friday, 25. June 2021 10:20AM
Finished at Friday, 25. June 2021 10:24AM
6055981 reads UMI processed
514 reads processed
158 families processed
    102 unrepresented families
    94 families with family size < 3
    6055467 families (DCS pairs) filtered for UMIs with mononucleotide repeats
    0 families filtered for Ns in the UMI
64 SSCS made
    0 SSCS filtered for excessive Ns
18 DCS made
    112 DCS failed due to missing SSCS
    2 DCS filtered for excessive Ns

Some of this is on me for not making these changes in a separate branch from the dev branch.

[bkohrn]

OK; I now believe this is giving the same output as v2.1.1.