Shrimp Short Reads Aligner

Question

[Joe]

Hey all, I've recently been introduced to short reads mapping and is planning to try on SHRiMP v2.2.0

Currently I have:-

36bp letter-spaced paired-end short reads from illumina and 16kb reference genome.

I would like to know if:-

1. SHRiMP allows the input of letter-spaced fastq files or is it just for color-spaced data?

2. Do we need to build any indexes for SHRimp prior to the alignment? I've seen some seeds option there but I am not too sure on how i should optimise it. Any suggestions there?

3. I've tried on this command

gmapper-ls -1 read1.fasta -2 read2.fasta reference.fasta -s w12 -E >map.reads.sam 2>map.read.log The SAM output file doesnt seemed to give me any results when I convert it to the VCF file with samtools.

Am I missing out any commands there?

Cheers and thansk in advance!

Answer 1

[Torst]

1. SHRiMP 2.2.1 is now out

2. Use gmapper-ls for letter-space (ATGC) and gmapper-cs for colour-space (0123). For letter-space it accepts FASTA and FASTQ.

3. You do not need to pre-build the index. SHRiMP will do it at run time.

4. Your command is faulty - you need to put all options BEFORE the reference genomes. Also, -E (--sam) is the default output mode, so you don't need it. And "-s w12" is not documented explicitly, and is the default anyway, so leave it out.

5. You have paired reads, but have not enabled --pair-mode (-p). For Illumina use "-p opp-in".

6. Look at the output SAM file with a text editor or "less" before further processing.

Answer 2

[Istvan Albert]

Answers to most of your questions can be found in the SHRiMP manual

• don't worry about the seeds yet, get your whole pipeline working first, you can revisit parameter tuning later
• check your samfile first with other tools like samtools view