<@UQWKJ27RQ> I'm nearly there with the racon modul...
# bacass
a
@Daniel Straub I'm nearly there with the racon module. The holdup now is that there isn't a genome paf file in the test data to test the module
d
Yes, I have seen that this morning as well. I have tried quick & dirty to fix it, but the
bacteroides_fragilis
dataset did not yield a working
paf
Could you do the following and see whether you can use a good
paf
?:
Copy code
wget https://depot.galaxyproject.org/singularity/minimap2:2.21--h5bf99c6_0
singularity shell minimap2:2.21--h5bf99c6_0
    minimap2 \
        -ax map-ont \
        -t 2 \
        test.fastq.gz \
        genome.fna.gz \
        > test_genome.paf
a
As we can uses the sars test data for the other inputs we could try generating a paf file from that data
d
Yes, any data will do, but as i said, for
bacteroides_fragilis
the produced data gave an error
Could you try that? If not, I can have a look
a
I'm trying the minimap2 test which should produce a test.paf file for the sars data
d
👍
@Andries van Tonder How is it going with the test data?
a
I was running into an error. I think I know what the problem is; working on it now
d
Alright, good to hear!
a
I added genome.paf to test-data and submitted racon for review
d
Great! I will have a look!
a
Is there a room for the bacterial genomics group?
d
No, I am currely in the lecture hall idling 😄
Sorry for the late review, added some comments
a
Ok, would you have time to have a quick chat say in an hour? I've been working on the miniasm module which also requires a paf file. There's an issue with the paf file I used for racon and the sequencedata I used for the mapping
d
Alright, any time in gather town.
a
Ok, just need to have some lunch
d
Sure
a
Ok, so it looks like we need two paf files: one generated by running minimap2 on the ont data twice (overlap) which miniasm is happy with; and one generated by running minimap2 on the genome fasta and ont data which racon is happy with. It still doesn't work with the sars test data but does with the bacteriodes test data so we should be able to use that
The issue I'm having now is that the awk one liner included in the bacass miniasm module is throwing a syntax error when I try running it on the gfa file generated by miniasm:
awk '/^S/{print ">"\$2"\\n"\$3}' test_ont.gfa | fold > test_assembly.fasta
Screenshot 2021-10-28 at 14.15.26.png
d
Sure, this is becasue it is escaping stuff (because in process script block) that need to be removed when using it in bash directly
Use
awk '/^S/{print ">"$2"\n"$3}' test_ont.gfa | fold > test_assembly.fasta
a
I tried removing a couple of the backslashes and got it to work
Ok, I made the changes to racon you suggested as well as updated the tests. Also I've submitted a PR for miniasm which needs reviewing
How do we get the racon module merged? Add a message on the request-review channel?
d
Hm I did approve the PR, but it might be the you cannot merge it yourself because one test is failing (conda md5). I would suggest writing this to the slack modules channel and ask somebody with write right (who can bypass test restrictions) to merge the PR