SOAPdenovo2 long run

Phoenix · 01-29-2020, 05:55 AM

Hello all,
I’m trying to assemble a plant genome (this is my first time). I have 100bp PE reads (106G forward and 106G reverse). I’m using soapdenovo2 to assemble the data on a cluster. The job was running on 10 nodes with each using 20 threads and 80Gb of memory. The maximum amount of time allowed for a job is 7 days.
I tried the following script however the job does not finish in the allotted time
SOAPdenovo-63mer all -s ./soap.config -K 63 -R -o genome

So then I tried to build it in parts using the script below, but even this does not finish
SOAPdenovo-63mer pregraph -s ./soap.config -K 63 -R -o genome

If using more computer resources or getting my time limit extended is not an option, how can I assemble the genome?

I tried using DMTCP to checkpoint the run so that I can resubmit the run and continue the assembly after the 7 days but I cannot get DMTCP to work with soapdenovo2.

I would be grateful for any assistance
Cheers

SNPsaurus · 01-30-2020, 04:26 PM

I'd try a speedy assembler like BBtools tadpole.sh as a first pass just to see what a rough draft looks like. A slower assembler but one that should finish in a day is abyss-pe with the bloom filter option enabled.

Phoenix · 01-31-2020, 11:01 PM

Thanks for the suggestions. I'll give these tools a try

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How long miso --run would take for a single BAM file	adrian	Bioinformatics	0	01-26-2017 12:57 PM
CNVnator histogram run for too long	CottusCNV	Bioinformatics	15	12-16-2016 01:11 AM
tophat taking too long to run	fahim	RNA Sequencing	7	05-14-2014 08:59 AM
Very long tophat run	seq_newbie	Bioinformatics	3	04-15-2014 08:02 AM
How long should paired-end alignment run?	agc	Bioinformatics	11	09-07-2011 01:31 AM

01-29-2020, 05:55 AM	#1
Phoenix Junior Member Location: UK Join Date: Aug 2013 Posts: 6	SOAPdenovo2 long run Hello all, I’m trying to assemble a plant genome (this is my first time). I have 100bp PE reads (106G forward and 106G reverse). I’m using soapdenovo2 to assemble the data on a cluster. The job was running on 10 nodes with each using 20 threads and 80Gb of memory. The maximum amount of time allowed for a job is 7 days. I tried the following script however the job does not finish in the allotted time SOAPdenovo-63mer all -s ./soap.config -K 63 -R -o genome So then I tried to build it in parts using the script below, but even this does not finish SOAPdenovo-63mer pregraph -s ./soap.config -K 63 -R -o genome If using more computer resources or getting my time limit extended is not an option, how can I assemble the genome? I tried using DMTCP to checkpoint the run so that I can resubmit the run and continue the assembly after the 7 days but I cannot get DMTCP to work with soapdenovo2. I would be grateful for any assistance Cheers

01-30-2020, 04:26 PM	#2
SNPsaurus Registered Vendor Location: Eugene, OR Join Date: May 2013 Posts: 521	I'd try a speedy assembler like BBtools tadpole.sh as a first pass just to see what a rough draft looks like. A slower assembler but one that should finish in a day is abyss-pe with the bloom filter option enabled. __________________ Providing nextRAD genotyping and PacBio sequencing services. http://snpsaurus.com

01-31-2020, 11:01 PM	#3
Phoenix Junior Member Location: UK Join Date: Aug 2013 Posts: 6	Thanks for the suggestions. I'll give these tools a try