I have to run all the sample with a single run

Question

[B- for bioinformatics]

Hello

Now, I am trying to do eukaryotic de novo genome assembly (Illumina paired-end samples). I have more than 25 samples (sequenced fastq files). I have arranged all 25 sequenced fastq files into a separate folder that rooted in one folder (Mother folder). >>>See below

                                            Mother folder

sample1                   sample2           sample3        sample4.........................................sample 25

I would like to run all the 25 sequenced data simultaneously until the annotation steps.

What could be the possible codes or expressions that I can use in the Linux platform, Does anyone has done it before?

[Ram]

What have you tried? This is a basic shell question (if you have a command that works for one sample). If you don't have such a command, you need to give us details on the software you intend to use, FASTQ file naming conventions, FASTQ file - sample name mapping patterns, etc so we can help you.

[B- for bioinformatics]

Sample: s.cerevisiae strains_ whole genome sequence data (I have sequenced data of 25 strains)

So, far I have tried a de novo assembly with one sample (2001_1 and 2001_2), which means I have generated all scripts for the de novo assembly.

what do I think?

How do I run scripts (the script I used before) parallel to 25 strains that are saved in different work directory, from Linux shell?

I can save time if I could find shell scripts for this?

[Ram]

• What tool did you use for the de novo assembly?
• What sort of parallel computing facilities do you have? A cluster? AWS? Just a laptop?
• What have you tried to learn yourself about running multiple commands in parallel on linux? Did you google run parallel shell commands?

[B- for bioinformatics]

fastqc, Trim_galore, kmergenie, assemblers (velveth and velvetg), mapping(BWA) and igv.

Just a laptop (1.4TB free space and 16gb RAM)

Yes, I did. I am going through it......

[Ram]

Try gnu parallel. You should be able to create a file with 25 commands and run each line using a separate core using parallel < commands.txt (which I figured out by googling gnu parallel commands from file). Google is your friend.

[lieven.sterck]

Bluntly I would say add them all in your commandline for the assembly tool, but as RamRS pointed out, without extra info (eg which tool do you plan to use, some might take a lof as input as well ) it's near impossible to provide you with a more detailed answer.

(lof == list of files )