Default consensus slack inconsistency

[ACastanza]

Is the bug primarily related to salmon (bulk mode) or alevin (single-cell mode)? Salmon (build mode)

Describe the bug A clear and concise description of what the bug is.

From salmon output: "validateMappings implies a default consensus slack of 0.2. Setting consensusSlack to 0.35"

When passing the --validateMappings flag without an explicit consensus slack, the program says that a 0.2 value is implied for the consensus salck, but then proceeds to set a different value.

To Reproduce Steps and data to reproduce the behavior: Pass --validateMappigngs without explicit consensusSlack Specifically, please provide at least the following information:

• Which version of salmon was used? - 1.3.0

• How was salmon installed (compiled, downloaded executable, through bioconda)? bioconda

• Which reference (e.g. transcriptome) was used? Mouse gencode M25

• Which read files were used? Internal experiment

• Which which program options were used?

  salmon quant
  --index=$index
  --threads=12
  --validateMappings
  --gcBias
  --posBias
  --seqBias
  --libType=A \ (ISR was detected, which is correct) -1 "${R1[$i]}"
  -2 "${R2[$i]}"
  --numBootstraps=30
  --output=$outdir ;

Expected behavior A clear and concise description of what you expected to happen.

The default consensusSlack and the value that is set when it is not explicitly provided should be the same value.

Desktop (please complete the following information): ProductName: Mac OS X ProductVersion: 10.15.7 BuildVersion: 19H2

[ACastanza]

This is unrelated, but didn't seem worth opening a whole separate issue for, there seems to be a misplaced newline character in the processed N fragments dialogue:

[rob-p]

Hi @ACastanza,

Thanks for reporting both of these. For the first, I think it is just the case that the message needs to be updated. In fact, the --validateMappings flag is now deprecated since selective-alignment is used by default (and can't be turned off, except in the single-cell mapping context, with the --sketchMode flag, which is currently only in the develop branch). We'll update that message. Regarding the misplaced newline, the issue is that the other messages are written by the logger, which is asynchronous. So, sometimes it will get to the appropriate place and write a newline before the fragment counter starts, and sometimes it won't. I'll look into if there is a way to better clear the line, even if the update is asynchronous.

Thanks!