[usad]

The ^M would indicate you have a windows file on a linux system.
Try dos2unix on the script and try again.

[a_crisp]

Using single-end reads

Hi,
Is it possible to use the program with single-end reads? Specifically Pacbio reads.
Thanks,
Alastair

[mht]

Hi boetsie,

I notice that GapFiller is introducing N's into my contigs where it previously has nucleotides. I've tried -t 0 but it still does not work. Any idea why this is and how this can be solved?

Thanks.
mh


Correction: my bad, it's an actual gap. sorry about this.

[yifangt]

IMAGE manual

Hello,
Can anyone share your experience using IMAGE for gap closing. I could not get it run after a couple of hours try. I need you explain the options for IMAGE command line for me,

/absolute_path_of_image/image.pl -scaffolds scaffolds.fa -prefix 76bp -iteration 1 -all_iteration 10 -dir_prefix iter

Particularly the -prefix option. From my understanding, it is the prefix of the paired end fastq file.
1) Should the fastq file be in the current directory? which did not work in my case.
2) What if the reads are fasta files which are used for my assembly?
Try to find the manual by writing to the author/group, no response. Googled a while, no luck. Tried PAGIT which is now the bundle including IMAGE, no instruction either.

Thanks a lot!

YT

[toddknutson]

I am having same problem. did you figure out how to fix it?

You need to supply Gap Filler with a single FASTA (not multi-fasta) file. Thus, you need to submit a file like this:


>all_contigs_linked_with_N's
ACGTACGT
ACGTACGT
ACNNNNN
NNNNNNN
NNNNNAC
GTACGTA


and NOT:
>contig1
ACTGT
>contig2
ACGTGACT

[toddknutson]

You need to supply Gap Filler with a single FASTA (not multi-fasta) file. Thus, you need to submit a file like this:


>all_contigs_linked_with_N's
ACGTACGT
ACGTACGT
ACNNNNN
NNNNNNN
NNNNNAC
GTACGTA


and NOT:
>contig1
ACTGT
>contig2
ACGTGACT

[Markiyan]

Can GapFiller accept known repeats sequences?

Can GapFiller accept know repeats sequences in multifasta format, and use those to improve repeats resolution?:

1. First map the known repeats to the assembly.
2. Avoid using reads anchored only in the repetitive region, and first rely only onto reads anchoring onto unique sequence flanking the repetitive regions.
3. Do a mini assembly of the gap caused by the repeat, if unable to close the gap de novo - use the repeat sequence as a base for consensus and map the mates/de novo miniassembly contigs to it. Correct the consensus, where contigs are of a sufficient quality (>Q35?) and have no high quality unaligned portions to the repeat consensus.

PS: While the pilon is quite a good tool, it has problems when used in the iterative mode (like IMAGE), that areas extending into the repeats attract all reads from the other repeat copies during the next mapping step, stalling the repeat extension/gap closures for the robed repeat copies.

PPS: Getting a repeats library for PCR-free datasets: get the median coverage for the assembly, and do subsambling with low coverage assembly (2x-4x) in order to get the repetitive sequences (check contig coverage for that).

[jgroh]

Does gapfiller create problems for downstream alignment

Hello,

I'm interested in using GapFiller, however I am wondering whether this might cause issues for downstream alignment. In particular, if the original assembly contigs comprise close to the entire genome, but they were not all able to be scaffolded, then will using GapFiller essentially cause these sequences to be duplicated in the final reference sequence? For instance if contig B was not able to be scaffolded between contig A and contig C, but paired reads can be used by GapFiller to fill in the sequence of contig B, this would mean that reads which originate from the sequence contained in contig B will map equivocally to both contig B and the scaffold in between contig A and contig C. Is this correct, and is not expected to create any problems such as genome inflation and equivocal mapping?

Jeff