Getting unassigned sequence IDs

[bruvellu]

Hello,

The code for getting the unassigned IDs in the tutorial does not seem right:

tail -n 1 test_human.partition.txt | tr ',' '\n'  > unassigned.list

# Outputs the last row of the final partition with members split by newline:
9_0	2	HQ_152
HQ_22

I used the command below. Probably not the cleanest, but works:

grep -R '#unassigned:' preCluster_out | cut -d':' -f3 | tr ',' '\n' | sort -V -u | sed '1d' > preCluster_out.unassigned.ids

These unassigned transcripts were not included in the preCluster_out.orphans.fasta. Is this the expected behaviour?

Thanks!
Bruno

[alissawilliams]

Thanks for pointing this out. I came across this post while trying to figure out issues with my unassigned sequences, and you're right—the command in the tutorial only gave me one sequence because tail -n 1 just takes the last line. I used your command and got 2,107 sequences.

However, I'm still confused about where all of my sequences are going. In my final_partition.txt file, there are 20,000 fewer sequences than there are in my input file. I've found 2,017 in preCluster_out.unassigned.ids, 3,719 in preCluster_out.orphans.fasta, and 14,244 in preCluster_out.tucked.fasta, but 118 sequences are still unaccounted for. I have tried to find specific transcripts that aren't in final_partition, but they aren't in any of the aforementioned files. Does anyone know why this is the case?

Thanks!