Which simulator to use for generating fastq reads from a population of haploids

Question

[Joseph Hughes]

Hi,

I want to generate reads from a set of different haplotype sequences found at different frequencies within a sample. I was wondering whether any Biostar users had any experience of doing this and which tools they used. I would prefer using an existing tool but there are quite few out there (I have compiled some in the table below). I am wondering which one will suit my purpose best.

Any advice is very welcome - thanks.

Joseph

Name                         Reference                                         Single-     Paired-     insert size         Quality           customise     coverage bias     Phred score from     Simulated sequencing     insertion-deletion     source                                            platforms
                                                                               end         end         between reads       score             read length                     existing FASTQ       errors                   errors

ArtificialFastqGenerator     doi: 10.1371/journal.pone.0049110                 -           yes         yes                 yes               yes           yes based         yes                  yes                      -                      https://sourceforge.net/projects/artfast​qgen/     illumina
                                                                                                                                                           on GC
ART                          doi: 10.1093/bioinformatics/btr708                yes         yes         -                   no                yes           -                 yes                  yes                      yes                    -                                                 Roche 454, Illumina Solexa, Applied Biosystem SOLiD
WgSim                        doi: 10.1093/bioinformatics/btp352                -           -           -                   dummy             -             -                 -                    uniform distribution     yes by simulating      -                                                 -
                                                                                                                           quality                                                                error                    INDEL polymorphisms
                                                                                                                           scores
Mason                        Holtgrewe M (2010) Mason - a read                 yes         yes         -                   yes               -             -                 no                   yes                      random                 -                                                 454, Illumina, Sanger
                             simulator for second generation
                             sequencing data. Technical report,
                             FU Berlin.
SimSeq                       Available: https://github.com/jstjohn/SimSeq.     yes         yes         yes                 -                 <100bp        -                 -                    -                        -                      -                                                 Illumina 
                             Accessed 2012 October 10th.
pIRS                         doi: 10.1093/bioinformatics/bts187                -           yes         -                   empirical model   -             yes based         -                    empirical model          yes with               -                                                 -
                                                                                                                           based on                        on GC                                  based on                 additional
                                                                                                                           read cycle                                                             read cycle               tool
Stampy                       doi: 10.1101/gr.111120.110                        -           -           -                   -                 -             -                 -                    -                        -                      -                                                 -    
MetaSim                      doi: 10.1371/journal.pone.0003373                 yes         -           -                   no                -             -                 -                    -                        -                      -                                                 -
FlowSim                      doi: 10.1093/bioinformatics/btq365                yes         -           -                   yes               -             -                 -                    -                        -                      yes                                               454
simNGS                       http://www.ebi.ac.uk/goldman-srv/simNGS/          -           -           -                   -                 -             -                 -                    -                        -                      -                                                 -
Grinder                      doi: 10.1093/nar/gks251                           -           -           -                   -                 -             -                 -                    -                        -                      -                                                 -

Answer 1

[Joseph Hughes]

I started off using ArtificialFastqGenerator but it first ran out of memory and then when I provided more memory it took a very long time to run.

In the end, I found that ART was the most useful program. At least I could get it to work without too much difficulty. I generated fastq reads from a number of different references at different depths of coverages and then mixed different sets together.

UPDATE: The following blog post is related and very useful - http://scottmyourstone.blogspot.co.uk/2013/10/read-simulators-review-with-emphasis-on.html

[Medhat]

Do not you think that ArtificialFastqGenerator excel over ART?! How long it take you to run ArtificialFastqGenerator? What was the specification of the computer that run it?