In Mainz, we believe that fastas are the future

Rationale

There ain't much to do with fastas but to download them and use. For this you could open the terminal (or powershell on windows) and type

wget -O human.fasta "http://www.uniprot.org/uniprot/?query=reviewed:yes+AND+organism:9606&format=fasta"

to download human database. Then, to get contaminants used in our group, you could simply type

wget -O contaminants.fasta "https://raw.githubusercontent.com/MatteoLacki/protein_contaminants/master/contaminants.fasta"

Finally, you would merge the two files as simple as

cat human.fasta contaminants.fasta > ready.fasta

and would feed ready.fasta into the software (on Windows you might use type instead of cat).

To do this in Python with furious_fastas you can simply run:

from furious_fastas import fastas, contaminants

fs = fastas("http://www.uniprot.org/uniprot/?query=reviewed:yes+AND+organism:9606&format=fasta")
fs.extend(contaminants)
fs.write('/my/favourit/location/for/fastas/human.fastas')

Need to add in reversed sequences? We have you covered:

fs.reverse()
fs.write('/my/favourit/location/for/fastas/human_rev.fastas')

Do you work with proteome mixtures? Here's how to deal with a search involving human, yeast, and ecoli:

from furious_fastas import Fastas, contaminants

fs = Fastas()
fs.download('http://www.uniprot.org/uniprot/?query=reviewed:yes+AND+organism:9606&format=fasta')
fs.download('http://www.uniprot.org/uniprot/?query=organism:643680&format=fasta')
fs.download('http://www.uniprot.org/uniprot/?query=reviewed:yes+AND+organism:83333&format=fasta')
fs.extend(contaminants)
fs.reverse()
fs.write('/my/favourit/location/for/fastas/hye.fastas')

Now, we also automate the process of managing the database with fasta folders, cause, you know, why not? To do this, use the update_fastas script.

% python3 update_fastas -h

usage: update_fastas [-h] db_path species2url_path

Update fasta databases.

positional arguments:
  db_path           Path to the folder treated as database.
  species2url_path  Path to the file mapping species names to urls.

optional arguments:
  -h, --help        show this help message and exit

However:

• do make sure that no software is running while we update the fastas
• do make sure that python is to be found in system variable Path
• add the whole thing to chronjobs or windows scheduler to make it automatic

Finally, it's sometimes good to simply parse fasta files for Python scripting. biopython is bloated, and we keep it much more free of additional dependencies.

Best Regards, Matteo Lacki