search

andreas-wilm / lofreq3

LoFreq Version 3
MIT License
27 stars 0 forks source link

Ready to replace lofreq2? #44

Open charlesfoster opened 2 years ago

charlesfoster commented 2 years ago

Hi,

This issue isn't really an issue per se, but more a question about the status of lofreq3.

I use lofreq version: 2.1.5 in an analysis pipeline for SARS-CoV-2. I like updating to the latest versions of tools etc. early on, but only if they're ready to replace the existing versions. So, is lofreq3 ready to replace the use of version 2.1.5, or is it still in more of a development phase?

Additionally, we use either nanopore sequencing or Illumina sequencing on a regular basis, depending on batch size or urgency. Is lofreq3 suitable for both platforms? Are there are necessary tweaks for either platform?

Thanks, Charles

charlesfoster commented 2 years ago

I should also now add that I just got an error with lofreq3 when trying to process a sample...

alias lofreq3=/path/to/lofreq
# prep BAM
samtools sort -u -@ 20 -n  $BAM | \
samtools fixmate - - | \
lofreq3 viterbi -f $REF -b - | \
samtools sort -u -@ 20 - | \
lofreq3 indelqual -f $REF -b - | \
lofreq3 alnqual -f $REF  -b - | \
samtools view -b - -o $OUT

lofreq call -b $OUT -f $REF

FATAL(lofreq_call.c|main_call:1261): Couldn't parse Bonferroni factor

Removing the -b flag made the 'call' command run. After post-processing the VCF file, I got the same variants as with lofreq2, just with very marginally different allele frequencies.