De novo assembly for virus genome with Velvet.
Dear expert,
I want to assembly a virus genome using velvet. The virus genome is about 3KB length. In the follow command, ref1.fa is the reference genome. But it seems that I can not success. I cannot get back ref1.fa from the simulate reads.
~/bin/bioinfomatics/wgsim-master/wgsim -N 500000 -1 100 -2 100 -h ~/projects/virus/analysis/ref1.fa r1.fq r2.fq
~/bin/bioinfomatics/velvet_1.2.10/contrib/shuffleSequences_fasta/shuffleSequences_fasta.pl r1.fq r2.fq output.fq
~/bin/bioinfomatics/velvet_1.2.10/contrib/VelvetOptimiser-2.2.4/VelvetOptimiser.pl\
-s 27 -e 31 -f '-longPaired -fastq output.fq' -t 4 --optFuncKmer 'n50'
Dec 1 17:18:33
Will run velvet optimiser with the following paramters:
Velveth parameter string:
-shortPaired -fastq output.fq
Velveth start hash values: 27
Velveth end hash value: 31
Velveth hash step value: 2
Velvetg minimum coverage cutoff to use: 0
Read tracking for final assembly off.
Dec 1 17:18:33
Beginning velveth runs.
********************************************************
Assembly id: 1
Velveth timestamp: Dec 1 2014 17:18:57
Velveth version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_27 27 -shortPaired -fastq output.fq
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_27
Velvet hash value: 27
Roadmap file size: 110519999
**********************************************************
********************************************************
Assembly id: 2
Velveth timestamp: Dec 1 2014 17:18:59
Velveth version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_29 29 -shortPaired -fastq output.fq
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_29
Velvet hash value: 29
Roadmap file size: 107811920
**********************************************************
********************************************************
Assembly id: 3
Velveth timestamp: Dec 1 2014 17:19:00
Velveth version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_31 31 -shortPaired -fastq output.fq
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_31
Velvet hash value: 31
Roadmap file size: 103399704
**********************************************************
Dec 1 17:19:00
Beginning vanilla velvetg runs.
********************************************************
Assembly id: 1
Assembly score: 53
Velveth timestamp: Dec 1 2014 17:18:57
Velvetg timestamp: Dec 1 2014 17:21:52
Velveth version: 1.2.10
Velvetg version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_27 27 -shortPaired -fastq output.fq
Velvetg parameter string: auto_data_27 -clean yes
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_27
Velvet hash value: 27
Roadmap file size: 110519999
Total number of contigs: 1062
n50: 53
length of longest contig: 95
Total bases in contigs: 58657
Number of contigs > 1k: 0
Total bases in contigs > 1k: 0
**********************************************************
********************************************************
Assembly id: 2
Assembly score: 57
Velveth timestamp: Dec 1 2014 17:18:59
Velvetg timestamp: Dec 1 2014 17:22:06
Velveth version: 1.2.10
Velvetg version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_29 29 -shortPaired -fastq output.fq
Velvetg parameter string: auto_data_29 -clean yes
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_29
Velvet hash value: 29
Roadmap file size: 107811920
Total number of contigs: 424
n50: 57
length of longest contig: 95
Total bases in contigs: 25334
Number of contigs > 1k: 0
Total bases in contigs > 1k: 0
**********************************************************
********************************************************
Assembly id: 3
Assembly score: 61
Velveth timestamp: Dec 1 2014 17:19:00
Velvetg timestamp: Dec 1 2014 17:22:07
Velveth version: 1.2.10
Velvetg version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_31 31 -shortPaired -fastq output.fq
Velvetg parameter string: auto_data_31 -clean yes
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_31
Velvet hash value: 31
Roadmap file size: 103399704
Total number of contigs: 1917
n50: 61
length of longest contig: 99
Total bases in contigs: 121773
Number of contigs > 1k: 0
Total bases in contigs > 1k: 0
**********************************************************
Dec 1 17:22:07 Best assembly by assembly score - assembly id: 3
Dec 1 17:22:07 Optimisation routine chosen for best assembly: shortPaired
Dec 1 17:22:07 Looking for the expected coverage
Dec 1 17:22:09 Expected coverage set to 0
********************************************************
Assembly id: 3
Assembly score: 61
Velveth timestamp: Dec 1 2014 17:19:00
Velvetg timestamp: Dec 1 2014 17:22:07
Velveth version: 1.2.10
Velvetg version: 1.2.10
Readfile(s): -shortPaired -fastq output.fq
Velveth parameter string: auto_data_31 31 -shortPaired -fastq output.fq
Velvetg parameter string: auto_data_31 -clean yes -exp_cov 0
Assembly directory: /Users/jhuang/projects/hbv/analysis/auto_data_31
Velvet hash value: 31
Roadmap file size: 103399704
Total number of contigs: 1917
n50: 61
length of longest contig: 99
Total bases in contigs: 121773
Number of contigs > 1k: 0
Total bases in contigs > 1k: 0
Paired Library insert stats:
**********************************************************
Dec 1 17:22:09 Setting the short insert length
Dec 1 17:22:09 Setting assembly short insert length(s) to auto
Dec 1 17:22:09 Beginning coverage cutoff optimisation
Minimum specified coverage cutoff is higher than the expected coverage. Please choose a minimum coverage cutoff smaller than 0 and re-run.
2 answers
Something appears to be going wrong before that error as none of your contigs are larger than 99 bases. A few points:
- Your input data type is 100bp paired end right? I wouldn't describe that as 'long paired end' in your command. But it seems that velvetoptimiser is running shortpaired anyway, so that may or may not be an issue.
- What does the quality look like on the data? I don't know wgsim, but could the artificial error profiles be screwing it up?
The coverage is too deep for velvet to handle, try to reduce the coverge to 50x - 100x. e.g. wgsim -N 1500.
With your current simulated data, use velvet-estimate-exp_cov.pl to evaluate the coverage-cutoff (say 300), add -exp_cov auto -cov_cutoff 300 parameters to velvetg
Try normalization the reads before assembly. http://ged.msu.edu/papers/2012-diginorm/
Log in to answer this question.
More posts like this
-
Using VelvetOptimiser.pl within a forloop
written by c.e.chong •
-
Issues building indexed .bam files with SAMtools
written by lleipnitz •
-
velvetg failed during the assembly
written by slimane.khayi •
-
Doubts using Velvet (transcriptome assembly) mainly in performance
written by pablo61991 •
-
Velvet - ins_length auto
written by Kenny •
-
Velvetg not producing output files
written by Pryce Michener •
-
MetaVelvet caught in an infinite loop?
written by slvt16 •
-
Velvetg error coverage=0
written by blanca •
-
VelvetOptimiser: optimum cov_cutoff, PE insert stats, 2 Insert Size Libraries, N's without scaffold…
written by drtamermansour •
-
How to run velvetoptimiser on paired-end FASTQ datasets
written by whatup •
Here is my reference.