Resquiggle issue | Reads unsuccessfully processed

[aswathishiju]

Hi, I am testing with DNA nanopore data, one is control and another is treated sample. I just wanted to know the incorporation of Uracils instead of Thymine in my treated sample. I used Rel606_Ecoli B_strain.fasta file as reference seqeunce for resquiggling. The command I used is : tombo resquiggle /single_fast5 /Rel606.fa --obs-per-base-filter 99:200 --q-score 7 --signal-matching-score 2 --corrected-group RawGenomeCorrected_001 --basecall-group Baseball_ID_001 --overwrite --processes 20

The output I got shows that 100% reads unsuccessfully processed and 61.6% alignment not produced. However I got an index file of size 523 mb. The output screenshot is attached herewith.

What does that output exactly mean? Would it imply that the reference sequence which has Ts and the treated sample fast5s which has Us couldn't be aligned? Or is it due to any filters that I applied for resquiggling? The output also suggests that "revert with tombo-filter "clear_filters".

Any suggestions or inputs are highly appreciated.

Thanks in advance.

[SycamoreLeaf]

Hi @aswathishiju .

Would it imply that the reference sequence which has Ts and the treated sample fast5s which has Us couldn't be aligned?

Yes. Your treated sample is probably so different from a normal sample that the resquiggling algorithm can't make heads or tails of it. Maybe a sample with less uracil would work better.

[SycamoreLeaf]

Or is it due to any filters that I applied for resquiggling? The output also suggests that "revert with tombo-filter "clear_filters".

Tombo only managed to resquiggle 34.0% of the reads, but the alignment on these reads was so bad that Tombo "filtered" them out. The data is still there; if you run tombo filter clear_filters you can use these resquiggled reads as input for other commands. With such a bad alignment, these reads will probably not tell you anything.

[woshiyangsi]

I met the same problem.Is that mean i should change T to U in the transcript reference?