Using a two contig reference as the main ref.fa and unicycler output as the query about ragtag HOT 5 CLOSED

Hi Grace,

Can you share the commands you used to run ragtag? It would be strange to go from 125 to 12 contigs during correction. Correction should yield at least as many contigs than the input, since correction is just the breaking of input contigs. If the flye assembly was used as the query and the unicycle assembly used as the reference, that would make more sense. If that is the case, I can give you some suggestions.

Thanks,
Mike

from ragtag.

HI Mike,

I used placed minimap2 in my path then
python3 ragtag.py correct flye.fasta unicyc.fasta and
python3 ragtag.py scaffold flye.fasta ragtag_output/unicyc.assembly.corrected.fasta.

from ragtag.

Hi Grace,

Thanks for sharing your data. The unicycler assembly you provided actually only had 75 contigs, and after running ragtag on my end, Bif1.unicyc.corrected.fasta had 76 contigs, indicating that a single contig ("3") was broken one time during correction.

It appears that scaffolding also generated 76 contigs and that is beacuse two of the 76 corrrespond exactly to the two flye contigs.

Unicycler contig "1" corresponds to flye "contig_1" and Unicycler contig "2" corresponds to flye "contig_2". The flye assembly is 5676900 bp long and if you look at ragtag.scaffolds.stats you can see that the two placed contigs yield 5652627 bp of placed sequence, so that looks good to me.

I am not sure why we are getting different results. Can you confirm that your query assembly has 75 contigs rather than 125?

Thanks

from ragtag.

Ok, I switched the flye and unicyc... my bad.

Now, I see that there are 75 contigs in my unicyc.fasta.
I still have the idea that the assembly did not work because the output said there are 2 ragtag and a bunch of other numbers. But the stats say there are 0 misplaced sequences. Thanks for your help.

contig_1_RagTag
contig_2_RagTag
10
11
12
13
14
15
16
17
18
19
20
21
22

from ragtag.

Hi there,

Here is what my stats file looks like:

placed_sequences  placed_bp  unplaced_sequences  unplaced_bp  gap_bp  gap_sequences
2                 5652627    74                  286548       0       0

So this says that 286548/(286548+5652627) = 4.8% of sequence is unplaced, which is usually pretty good. If you are curious about why those other contigs were not placed, you can look at the intermediate query_against_ref.paf file to see what the alignments for the unplaced contigs look like.

I will close this for now but feel free to reopen if anything comes up.

Thanks

from ragtag.