Consensus sequences

[ceromova]

Hello,

I would like to ask if the consensus sequences of delly's output are human or just unknown insertion sequences (maybe virus or bacteria ones).

Thank you.

[tobiasrausch]

Hi,

The consensus sequences are computed by local assembly of all SV-supporting reads. They allow you to examine the SV breakpoint for homology patterns or small insertions.

Best, Tobias

[ceromova]

Therefore, if a read is half human genome half viral genome that would be detected by delly and report it as an insertion? This half-half sequence would appear in the consensus sequence?

Thank you.

[tobiasrausch]

For long reads, yes.

For short reads, the length of the reads is usually not sufficient and that's why people map against "augmented" references that include the viral sequences.

[ceromova]

What do you mean when you say long/short reads? Are you refering to Nanopore as a long read and Illumina as a short read?

[tobiasrausch]

Yes.