Feature request: Example workflow

[utimcraig]

Hi,
I'm new to bioinformatics and I am interested to use prospr to predict the structure of my protein of interest. I'd like to use this space to figure out how to actually run this software.

Step 1) Download and install the docker for prospr. This is already covered in the readme so I won't talk more about that.

Step2) Pick your amino acid sequence. Here is mine in FASTA format:

sp|P29279|CCN2_HUMAN CCN family member 2 OS=Homo sapiens OX=9606 GN=CCN2 PE=1 SV=2
MTAASMGPVRVAFVVLLALCSRPAVGQNCSGPCRCPDEPAPRCPAGVSLVLDGCGCCRVC
AKQLGELCTERDPCDPHKGLFCHFGSPANRKIGVCTAKDGAPCIFGGTVYRSGESFQSSC
KYQCTCLDGAVGCMPLCSMDVRLPSPDCPFPRRVKLPGKCCEEWVCDEPKDQTVVGPALA
AYRLEDTFGPDPTMIRANCLVQTTEWSACSKTCGMGISTRVTNDNASCRLEKQSRLCMVR
PCEADLEENIKKGKKCIRTPKISKPIKFELSGCTSMKTYRAKFCGVCTDGRCCTPHRTTT
LPVEFKCPDGEVMKKNMMFIKTCACHYNCPGDNDIFESLYYRKMYGDMA

Step 3) Generate a multiple sequence alignment from blastp I selected to blast against the non-redundant database.

Step 4) I got the results of the blastp, clicked on multiple sequence alignment.

Step 5) Clicked on download, the file I received is a *****.aln file. Now we need to change this alignment file into an a3m format file that prospr uses.

Step 6) From here I'm quite a bit more unsure of how to proceed. Anyone have advice?

I know I need to generate a multiple sequence alignment.

[maudagnotto]

You can use the hh-suite and convert the aln in a3m or directly generate the a3m as follow:
hhblits -cpu 3 -i query_T0954.seq -d /hh-suite/databases/UniRef30_2020_02 -oa3m query_T0954.a3m

You can find more information here:
https://github.com/soedinglab/hh-suite/wiki

However, I was not able to run prosPr with the option --a3mfile and a ticket indeed has been open.
Please let me know in case you manage to run the code with this flag.

Thanks :)