One thing that would make the documentation even more useful is to generate a bunch of stats from data across different organisms and platforms. And published that with the examples/docs.
In essence, the documentation could serve as a reference for choosing an aligner.
How long will it take to run aligner A on data obtained from organism B with coverage C and sequencing platform X. How good is that alignment? Are there any notable differences between aligners? Can I run that test myself - most of the time hunting down the data is the tedious part.
Currently, we are always guessing and usually go with past experience...
Maybe you would be interested in adding LAST to the collection of aligners? You can see in https://nf-co.re/pairgenomealign how I run in a whole-genome alignment pipeline.
I standardize the output for all aligner in sorted bam file. It is why it is specific to “read” alignment. I don’t think we can output bam with last. It exists already the pairgenomealign nf-core pipeline for whole genome alignment.
Indeed, LAST outputs to MAF. However it also provides a
maf-convertcommand that does SAM, whose output can be piped insamtools viewfor BAM or CRAM output. LAST has been used to align Nanopore RNA reads to genomes, for instance.Then yes it is feasible. Thx for the information
Lastis now present in AliNe v1.3.0Before that
bwa-mem2andsalmonhave been added in AliNe v1.2.0