Fungal Genome Assembly... a painful experience

[nwfungi]

Hi all and thanks for any insight you have in advance.

I am trying to assemble a fungal genome of approximately 42Mb in size. I have reads from an Ion Torrent run and some old data our lab had from a PacBio sequencing in fastq format.

I was wondering what would be a good assembly program to start attempting. I have heard that SPAdes is an option though the genome size is larger than bacteria. Falcon has also been suggested along with PBcR.

Any hints and help would be extremely appreciated.

[GenoMax]

Have you tried falcon as was suggested in a previous thread you had started? SPAdes is not going to work for this size genome.

BTW: How much PacBio data do you have and is it clean/long reads. If you don't have much/good data this is going to be a difficult task.

If you have a relatively close genome available then going the mapping route first may be useful. It will help you understand quality of your data.

[akorobeynikov]

Have you tried falcon as was suggested in a previous thread you had started? SPAdes is not going to work for this size genome.

This is certainly not true.

Any hints and help would be extremely appreciated.

I'd suggest to give SPAdes a try. If uncertain - just contact SPAdes support for some advices.

[GenoMax]

This is certainly not true.

I'd suggest to give SPAdes a try. If uncertain - just contact SPAdes support for some advices.

Has something changed in recent past?

We have seen the recommendation that SPAdes was designed for "standard isolates and single-cell MDA bacteria assemblies". There is no data on SPAdes site that shows successful assemblies with genomes of larger size/multiple chromosomes.

Can you provide some guidance on the RAM requirements for a genome this size? (It would depend to some extent on how much data @nwfungi has)

[pmiguel]

ABySS-PE seems to love assembling compact fungal genomes. That said we typically do PCR-free PE libraries + a cheap mate-pair library using only Illumina reads. N50>1 megabase and ~1000 scaffolds >1Kb total, is what I remember getting.

I haven't even seen an Ion Torrent data set, though, so I have no idea how that will work.

--
Phillip

[Bukowski]

Have you tried falcon as was suggested in a previous thread you had started? SPAdes is not going to work for this size genome.

BTW: How much PacBio data do you have and is it clean/long reads. If you don't have much/good data this is going to be a difficult task.

The PacBio coverage isn't the only critical question here (although the OP would do well to answer it!), but they suggest they have FASTQ files for the PacBio reads. Not sure about Falcon, but pretty sure HGAP.3 requires bax/bas.h5 as assembly input.

[AntonioRFranco]

Take a look to the Masurca assembler

http://www.ncbi.nlm.nih.gov/pubmed/23990416

[nwfungi]

Thanks for all the information and apologies for slow response time.

As for the PB data I have, it is in fastq format and we are in the process of trying to get the original files but this sequencing was done far enough in the past that it may not be possible. The quality is questionable and the coverage is low at best. I've only run FastQC on it to check it but nothing was flagged. We are having a more comprehensive PB sequencing effort done as we speak. I was more interested in seeing if the small amount of PB data I had could be used in the meantime to create a slightly more useful assembly and get a process hammered out.