Are Illumina reads still supported?

[osowiecki]

What about mate-pair illumina reads?
How do polish my pacbio restults?

[rvaser]

Hello,
Illumina reads for polishing (as in Ra) are not supported for now, as the underlying mapper has not yet been tested for mapping Illumina reads to the assembly. You can always use minimap2 + racon as a separate step after Raven assembly.

Best regards,
Robert

[osowiecki]

Assuming I want to do exactly that, probably based on that thread
isovic/racon#146
Should I set polishing rounds to 0 in raven or will it only help my pacbio assembly to keep "-p" value on default "2"?

[rvaser]

Use raven in default mode (so two iterations of PacBio polishing) and then follow the instructions for Illumina polishing as in the thread you listed :)

[osowiecki]

Thank you. I'm using quality trimmed illumina reads paired end + mate pairs.
Do you think using make pair reads with large insert sizes (smushed together with PE reads) are useful for polishing? How will racon treat reads with 11k insert size?

[rvaser]

Currently racon does not use the insert information, it treats both reads of a pair as single end reads. Thus I think nothing will happen if you merge together the PE and MP reads.

[osowiecki]

That clears things up.

As for the racon input. Does it really matter if R1 and R2 reads are put right next to each other in the shuffled fasta file?
The proposed "shuffle_pairs_fastq.py" produces reads with /11 as their name. Shouldn't pairs be named /1 and /2 or it doesn't really matter?

FCC4U5HACXX:1:1101:6258:2807#ACTTGAAT/11
ACGCGCTTCCCGGGAATGATCCCATGAAAAGGCGTCGCGCCTGCCCGGACCGTGGACAGATCGATCCGCAGGAGCCCGAGGGTCTCGGCGTAGATGAT
FCC4U5HACXX:1:1101:6449:2827#ACTTGAAT/11
GTTAGGCTGCCTCCAGTAAGACCCGCTAAAACGCACTGTCCGGTAAATATAGCGGATGCTATGCTCTTCTATGCCTCCAGCAACAACCGGTAAAACGT

[EDIT]

it treats both reads of a pair as single end reads
This is my answer, right?

[rvaser]

It does not matter, you can shuffle them as you like. All reads just need to have unique identifiers up to the first space.