I'm currently working on testing some matching algorithms for strings. I would like to do some tests on raw genomic data as I expect different results from random strings given the lower entropy. How can I get a whole human genome sequence? (just the raw basis)

I have tried websites like https://www.ncbi.nlm.nih.gov/ it seems very confusing and full of unnecessary information given that I just want to have the raw basis without annotations.

  • $\begingroup$ Its potentially an interesting project. I'm aware of comparable projects, but ... your question and investigation is ill-defined. Simply if you download 'fasta' format for any given sequence you will get the description on one line beginning with '>' and the sequence on the line below. Genbank is a good place to go $\endgroup$
    – M__
    Nov 5, 2020 at 22:50
  • $\begingroup$ @Michael I'd be happy to have references to the comparable projects, thanks! $\endgroup$
    – gcorso
    Nov 6, 2020 at 14:58
  • 1
    $\begingroup$ Just a small comment, "raw" usually in genomics means unprocessed sequencing reads: en.wikipedia.org/wiki/Read_(biology), the sequence you seem to be after is "genome assembly". $\endgroup$ Nov 8, 2020 at 13:27

1 Answer 1


You want FASTA files from here.

If you want the entirety of the genome, get the dna-toplevel file. Files with rm and sm have repeat sequences masked, the former by replacing the repeats with Ns (hard mask) the latter by designating repeats as lower case letters (soft masked).

See the README for more info.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.