A community effort to collect a curated set of analysis pipelines built using Nextflow.

nf-core

I just noticed <https://github.com/MultiQC/MultiQC/pull/2814|that MultiQC now has a module for UMI Collapse>! Great news, I wanted to write one for months and never got around to actually do it, so very happy - particularly since the code is much cleaner than what I would have managed to come up with. :smile:

However, since it was merged so quickly, I did not have the chance to point one little, but possibly important issue out: The search pattern is now defined as

```umicollapse:       
fn: "*_UMICollapse.log"       
num_lines: 100      
contents: "UMI collapsing finished in"```
Mind though, that the `*_UMICollapse.log`  name <https://github.com/nf-core/modules/blob/master/modules/nf-core/umicollapse/main.nf|is only a convention of the nf-core module>:

```UMICOLLAPSE_JAR=\$(dirname \$(which umicollapse))/../share/umicollapse-${VERSION}/umicollapse.jar
    java \\
        -Xmx${max_heap_size_mega}M \\
        -Xss${max_stack_size_mega}M \\
        -jar \$UMICOLLAPSE_JAR \\
        $mode \\
        -i ${input} \\
        -o ${prefix}.${extension} \\
        $args | tee ${prefix}_UMICollapse.log```
since the tool itself is not outputting any log file. Hence I was wondering if that search pattern should be applied or if it should be file contents only?

You're right, it should be file contents only

Would you mind putting in an issue / PR about it please? :pray::skin-tone-2:

Done, but I am unsure if the unique reads in FastQ mode are actually the dedup input reads of the BAM mode. Maybe <@UE4K7EZ4G> or <@U011VU477V0> knows more?
<https://github.com/MultiQC/MultiQC/pull/2827>

Dont know myself either, sorry <@U02ENQANGF9>