Anybody Has Experience In This Tool :Jellyfish? Need Help

Question

[GAO Yang]

Hi, I was trying Jellyfish on some downloaded data: Illumina-Ecoli100PE and the data sequenced in our lab. But none of them showed the curve like the first link do. All of mine results were the same: just downward, no peak (the 1-multiplicity had the most k-mers), no matter what K-mer used (tried 13,17,21,27,31). And because the current version supports fasta and fastq format, so I used this E.coli example directly as input file.

I believed results like this won't do any help to my purpose (find the peak, calc the coverage, and finally estimate the genome size) .

So, any of you have used this tool before? Could you share some exp? Maybe this tool just work normally for some certain kind of seq data? Thanks for your kindly help :)

Answer 1

[Vikas Bansal]

I have not used Jellyfish but I looked at it just now. I think it really depends on your data, species etc. If most of your k-mers have 1 multiplicity, I think it simply means that most of your reads are unique. As you said "none of them showed the curve like the first link do" , I think you should not compare your plot with those given in the link because your data is completely different from them. Just an example- consider you have fasta file of 25bp long reads and all of the reads are different from each other and then if you use Jellyfish with 25 k-mer, you will not get any peak. You will get just a line over 1-multiplicity. So I think it really depends on your data.