MetaVelvet crashes, no error given
Hello everyone,
I'm currently experimenting with different assembly programs with a small metagenomic dataset (~4 million reads) and I would like to try out Metavelvet but it seems to crash on me when I use the whole dataset.
I'm using Illumina paired end reads that are 251 bp long from a library prepared using NexteraXT. I have ~2 million forward and 2 million reverse reads so not much considering its a metagenomic data set.
I've tried out metavelvet using both the developer's small sample dataset as well as 500,000 of my own reads and they both assemble just fine. However, when I try to use the entire dataset it crashes at steps after using the meta-velvetg command.
The program is installed on a cluster and I'm using a high memory node to run the program.
Here are the commands I use:
velveth outdir_metav 47 -fasta -shortPaired R1.fasta R2.fasta
velvetg outdir_metav -exp_cov auto -ins_length 251;
meta-velvetg outdir_metav -ins_length 251 | tee logfile;
I understand that the higher the kmer size you use, the more memory the program requires to run so I've also tried re-running the program with different kmer sizes (35, 47, 53, and 61) and all of them seem crash.
I also tried to pair the reads and then split the file up such that every input file had fewer than 1 million reads in it. That also didn't work.
I've looked through forums and emailed the developers and have had no luck finding out what the issue is. I even had the program installed!
If anyone has any advice I would appreciate it!
I'm currently experimenting with different assembly programs with a small metagenomic dataset (~4 million reads) and I would like to try out Metavelvet but it seems to crash on me when I use the whole dataset.
I'm using Illumina paired end reads that are 251 bp long from a library prepared using NexteraXT. I have ~2 million forward and 2 million reverse reads so not much considering its a metagenomic data set.
I've tried out metavelvet using both the developer's small sample dataset as well as 500,000 of my own reads and they both assemble just fine. However, when I try to use the entire dataset it crashes at steps after using the meta-velvetg command.
The program is installed on a cluster and I'm using a high memory node to run the program.
Here are the commands I use:
velveth outdir_metav 47 -fasta -shortPaired R1.fasta R2.fasta
velvetg outdir_metav -exp_cov auto -ins_length 251;
meta-velvetg outdir_metav -ins_length 251 | tee logfile;
I understand that the higher the kmer size you use, the more memory the program requires to run so I've also tried re-running the program with different kmer sizes (35, 47, 53, and 61) and all of them seem crash.
I also tried to pair the reads and then split the file up such that every input file had fewer than 1 million reads in it. That also didn't work.
I've looked through forums and emailed the developers and have had no luck finding out what the issue is. I even had the program installed!
If anyone has any advice I would appreciate it!