How to choose a reference genome when there's no genome information in database for my samples?

Question

[Xiaokang]

I'm working on the RNA-Seq of cod fish. To align the seqs to a genome, I need to find a reference genome from database, such as UCSC or Ensembl. But unfortunately, there's no genome for cod fish. Should I then choose another genome, such human's, or there's other better option?

[h.mon]

The genome you want is Gadus morhua? I found it at ENSEMBL.

[Xiaokang]

Thank you. Too careless to find this.

Answer 1

[Nicolas Rosewick]

You could use different approaches :

1. Use a related species available genome (an other type of cod fish ?)
2. Use a de-novo assembly approach (e.g. Trinity)
3. Use a pseudo-alignment approach (e.g. kallisto) using cDNA sequences from the cod fish (if available) or an another related species.

edit : Or better : use the cod fish genome (Gadus Morhua) as suggested by h.mon http://www.ensembl.org/Gadus_morhua/Info/Index

Answer 2

[Tom_L]

Do not use a genome that is not related to your experiments, Ever.

How about: http://www.ensembl.org/Gadus_morhua/Info/Index (click on "Download DNA sequence")?