Hi, every one, Are there someone who have used reads simulator to simulate metagenomic reads, such as MetaSim. I met some problem in using Metasim. I want to simulate short pair end reads. But I don't know how to set insert length, read length and abundance value.
So, I recently am using a software named dwgsim that I saw it in MetaVelvet. The author who design MetaVelvet use it to simulate reads. I find it is easy to use.
I use dwgsim to simulate a metagenomic dataset from four species, NC_008525, NC_008527, NC_008700, NC_008095. Because my computer have only 2G ram, I only get the top 490,000 bp of every speice as the input of dwgsim, and I set read length 80 bp, insert length 500, error rate 1%. I use the all of the argument which is the same as MetaVelvet's paper. I set the abundance value is 60% for NC_008527, 10% for NC_008525, 15% for NC_008700, 15% for NC_008095, all of abundance value is set by arbitrarily. I just want to test some assebler performance. I generate three datasets by sequencing depth, 80x, 100x, 100x. Here, sequencing depth is all of reads bp is divided by 490,000*4(all species length). I used velvet and calculated the sum bp of correct contig(correct contig is identity > 95% and coverage rate(HSP is divided by the contig length) > 95% when contig is mapped to reference sequece), and then, the sum bp of correct contigs is divided by the length of the correspond species. I say it is the source sequence coverage rate.
To my surprised, this is my statistical data.
NC_008527(60%) NC_008525(10%) NC_008700(15%) NC_008095(15%)
80x 0.1265 0.03478 0.0281 0.00548
100x 0.1058 0.0277 0.0027 0.003
150X 0.0939 0.0036 0.0078 0.00337
As the increasing of sequecing depth, the source sequence coverage rate get down.
Where is the problem? I am going to changed a simulator, named NeSSm that seem to be used easily.
And I have another questhion, are there some body who have seen the paper <use of simulated data sets to evaluate the fidelity of metagenoimic processing methods>. The author just give the Organisms' name. I also know the Organisms' taxid. I don't know which Organisms I should download.
I look forward from you. Thank your very much.
your sicerely,
Yue Xu
So, I recently am using a software named dwgsim that I saw it in MetaVelvet. The author who design MetaVelvet use it to simulate reads. I find it is easy to use.
I use dwgsim to simulate a metagenomic dataset from four species, NC_008525, NC_008527, NC_008700, NC_008095. Because my computer have only 2G ram, I only get the top 490,000 bp of every speice as the input of dwgsim, and I set read length 80 bp, insert length 500, error rate 1%. I use the all of the argument which is the same as MetaVelvet's paper. I set the abundance value is 60% for NC_008527, 10% for NC_008525, 15% for NC_008700, 15% for NC_008095, all of abundance value is set by arbitrarily. I just want to test some assebler performance. I generate three datasets by sequencing depth, 80x, 100x, 100x. Here, sequencing depth is all of reads bp is divided by 490,000*4(all species length). I used velvet and calculated the sum bp of correct contig(correct contig is identity > 95% and coverage rate(HSP is divided by the contig length) > 95% when contig is mapped to reference sequece), and then, the sum bp of correct contigs is divided by the length of the correspond species. I say it is the source sequence coverage rate.
To my surprised, this is my statistical data.
NC_008527(60%) NC_008525(10%) NC_008700(15%) NC_008095(15%)
80x 0.1265 0.03478 0.0281 0.00548
100x 0.1058 0.0277 0.0027 0.003
150X 0.0939 0.0036 0.0078 0.00337
As the increasing of sequecing depth, the source sequence coverage rate get down.
Where is the problem? I am going to changed a simulator, named NeSSm that seem to be used easily.
And I have another questhion, are there some body who have seen the paper <use of simulated data sets to evaluate the fidelity of metagenoimic processing methods>. The author just give the Organisms' name. I also know the Organisms' taxid. I don't know which Organisms I should download.
I look forward from you. Thank your very much.
your sicerely,
Yue Xu