Simulating sanger sequence reads with mason

Question

[africys]

I am trying to simulate sanger sequencing reads with mason. The idea is to used mason to generate random snps based on a provided error.

Command

 mason sanger --read-length-error 0.1 test.fast

I am not sure what I am doing wrong but i get this error message

reference file: test.fasta
Loading reference sequence from "test.fasta"
Simulating for haplotype #0...
  Building haplotype...
    contig # 1/1    1.12    SNPs:0    indels:0    indel len sum:    0    vrate:0
  Finished haplotype creation in 0.000380993s
  Simulating reads for haplotype #0...
ERROR: haplotype (== 366) < read length!

What does this error mean? I could not find any documentation and description of mason error messages anywhere. Have you used mason

Answer 1

[Brian Bushnell]

Unfortunately, I have little experience with Sanger data. The BBMap package has a read simulation tool, "RandomReads", which will generate synthetic reads of specified length ranges with specified error rates, paired or single-ended. It only has quality profiles for Illumina and PacBio, but if those are not sufficient, I would be happy to add a Sanger profile.

A typical command is something like this:

randomreads.sh ref=hg19.fa snprate=0.20 maxsnps=3 delrate=0.01 \
               maxdels=1 maxdellen=5 insrate=0.01 nrate=0.01 \
               minlen=200 maxlen=800 reads=100000 maxq=40 \
               midq=35 minq=25 out=reads.fq

...which would generate 100000 reads with lengths between 200 and 800bp. 20% would have at least 1 SNP; 20% of those 20% would have 2+ SNPs, etc. up to a max of 3, and various other mutations as listed. Each read's ID is its original unmutated genomic location so that you can determine absolutely where a read came from. It's highly customizable, and explains most of the details if you run the shell script with no arguments, but let me know if you have questions.