Questions about Uniprot ID Mapping script

[yaaminiv]

Starting this discussion for questions about @kubu4's workflow: https://nbviewer.jupyter.org/github/RobertsLab/code/blob/master/notebooks/sam/20210601_ssal_gff-annotations.ipynb

I'm running the Uniprot mapping in this Jupyter notebook with this perl script. I got the following error:

Can't locate LWP/UserAgent.pm in @INC (you may need to install the LWP::UserAgent module) (@INC contains: /Users/Shared/Apps/anaconda/lib/perl5/site_perl/5.22.0/darwin-thread-multi-2level /Users/Shared/Apps/anaconda/lib/perl5/site_perl/5.22.0 /Users/Shared/Apps/anaconda/lib/perl5/5.22.0/darwin-thread-multi-2level /Users/Shared/Apps/anaconda/lib/perl5/5.22.0 .) at ../../code/Haws/08-uniprot_mapping.pl line 3.
BEGIN failed--compilation aborted at ../../code/Haws/08-uniprot_mapping.pl line 3.

What is LWP::UserAgent module and how do I install it?

[kubu4]

perl -MCPAN -e 'install Bundle::LWP'

You may need to use sudo. Try without sudo first, though.

[yaaminiv]

I ran perl -MCPAN -e 'install Bundle::LWP' (needed to use sudo). Needed to also run sudo cpan install LWP::Protocol::https (based on suggestion from this Stack Overflow post) since I got the error below:

Failed, got 501 Protocol scheme 'https' is not supported (LWP::Protocol::https not installed) for https://www.uniprot.org/uploadlists/

[yaaminiv]

After perl module installation and running the code, I get a sever closed connection error:

Failed, got 500 Server closed connection without sending any data back for https://www.uniprot.org/uploadlists/

How do I access the server?

[kubu4]

Hmm, try submitting again?

[yaaminiv]

Tried again and got the same error message/0 hits!

Line count:
       0 20210602_cgigas_roslin_uniprot_batch_results.txt
Failed, got 500 Server closed connection without sending any data back for https://www.uniprot.org/uploadlists/

This is my notebook: https://github.com/RobertsLab/project-oyster-oa/blob/master/code/Haws/08-GOterm-Annotation.ipynb

And this is my perl script: https://github.com/RobertsLab/project-oyster-oa/blob/master/code/Haws/08-uniprot_mapping.pl

Is there something in my perl script that isn't lining up?

[kubu4]

Well, I just spot-checked a couple of the IDs listed in your file via the UniProt website and there are no matching entries...

I'm wondering if the genome is new enough that the entries haven't been cross-referenced by UniProt yet?

[yaaminiv]

I'm wondering if the genome is new enough that the entries haven't been cross-referenced by UniProt yet?

Well shoot. Any other ideas for how to get GOterms using existing annotations? In this comment @sr320 suggested using protein IDs with UniProt mapping...is there a way to modify the current perl script I'm using for that purpose (assuming I spot check first and I'm getting matches)

edited to include the rest of my question (comment too quickly)

[kubu4]

I'm wondering if the genome is new enough that the entries haven't been cross-referenced by UniProt yet?

https://www.uniprot.org/help/synchronization

Looks like that might be the case:

However, UniProtKB data is "frozen" significantly earlier than the public release date, which can result in a delay of 8 to 16 weeks between the update of a related database and the appearance of relevant data in the public version of UniProtKB.

[yaaminiv]

Looks like using the protein IDs from the protein annotation FastA in Uniprot gives results! I'm going to try working from that file and isolate the IDs to see if that works. If there's a way to modify the perl script that's great, but if not I'll just use the GUI for now.

[kubu4]

In the Perl script, you'd want to replace P_ENTREZGENEID with P_REFSEQ_AC.

To get the list of IDs to submit, you can do this:

grep "^>" genome.fasta | awk '{print $1}' | sed 's/^>//' >> protein_list.txt

[yaaminiv]

To get the list of IDs to submit, you can do this: grep "^>" genome.fasta | awk '{print $1}' | sed 's/^>//' >> protein_list.txt

To clarify, I'd work from the protein FastA (file head below)?

(base) yaaminivenkataraman@big-wave ~ % head /Users/yaaminivenkataraman/Downloads/GCF_902806645.1_cgigas_uk_roslin_v1_protein.faa
>NP_001292217.1 uncharacterized LOC105326593 precursor [Crassostrea gigas]
MKVVPSLFIFIICVFISSGQRPDPFPKGCKAQIDVVFVIDGSDSISESDFQTLRTSISRIVDGFHIGSGETRMGIIVYSK
GVAFSVPLSYDPVYLKDQASIMPHPREGTNTHLGIEEMIDMFKKDKRDGVPMAGVVVTDGISKEKEKTLLQSRLARDLGI
NMFSVGVGRYTEEEELRGIASNPDQAIKVESFDELLKILSKLVQLVCPNKCMMPGVIAYPNDVSKNCRLYWKCEGEESKL
TCCPRGFSFSAPVQSCIPDPKCVEPCGDEGPICNKRPSVYQPTIYEELIEGYGWVQRSCPPGTAYDRVTCGCTITQTPPP
PRRVCRVLVHIPFDIDCVDTSGNGFLIKNHGVKFTRTGLALFEGKAKLVIPNIQRYLGSNFLVKMRYKEFPSFETQGLLS
NGDCYTPMTLQLIKDSIRHTYKIENSYRQRTGFTIPTNFKPWNEVTFSHDGNTLAGSVNGIQNEKRSNGPIQRTPYGISI
GFVDGYKHFQGYMDYVTIYKCANMNVYNKY
>NP_001292218.1 prostaglandin G/H synthase 2-like precursor [Crassostrea gigas]
MWWRAVLLVSFIAPALVSGGHNPCCSYPCENNGVCMTQGGNYYCDCANLDFYGKHCEIPTFMKKIKLFLKPSQSTLHNLL

[kubu4]

Yep. See my other suggestion about working from the GFF, as an alternative?.

[yaaminiv]

I'm going to stop commenting here and try your other suggestion! Using the protein IDs from the protein FastA, I still got 0 results:

Failed, got 500 Server closed connection without sending any data back for https://www.uniprot.org/uploadlists/