BiG-SCAPE db location error and GCF_id issue

[ramanujamr]

Hi

I'm running into some issues with nplinker usage and outputs.
I have 2 issues, first one I've tried to fix, but not sure if it lead to errors downstream.

Background:

Running NPLinker 2.0.0-beta.1 on Ubuntu 18.04.6.

This is the NPLinker toml file I used (renamed to .txt to be able to upload here).
nplinker.txt

And this is the python script I used:
run_nplinker.txt

Errors:

1. I get the following file not found error for big scape:
   FileNotFoundError: [Errno 2] No such file or directory: '/data/sdc/ram/20250715_DFIUM038/neg_mode/bigscape/bigscape_running_output/data_sqlite.db'
   I noticed that the bigscape output created was bigscape_running_output.db.
   To fix this manually, I renamed it to `data_sqlite.db'.
   When I reran the script, it worked and produced outputs (however, I'm not sure if it's the right way).
   bigscape_error.txt

2. When I look at the genomics_data.tsv from npl.to_tsv(link_graph) output, I have multiple comma separated GCF_ids in a single row linked to one product in a given strain_id. As I understand, it should be a unique GCF_id per row right?
   I'm not sure if something went wrong or if I'm not understanding this correctly.
   genomics_data.tsv

-Ram

[adraismawur]

Hey! Indeed, we changed the way that the database file is named, but I forgot to update NPLinker to accomodate this change. I will try to get a fix for this out soon.

[ramanujamr]

Would renaming and relocating be the only fix required?
I did that and ran NPLinker but I wasn't sure if that caused other unknown issues downstream.

[adraismawur]

That should work fine! I'll try to get a PR out for this issue this week.

[ramanujamr]

I see, thanks!
Then it looks like the multiple GCF_ids issue isn't linked to this.
Do you have any thoughts on where it might be coming from?

[justinjjvanderhooft]

The tsv file you shared, where is that again coming from? I am sure if this is a BiG-SCAPE thing or NPLinker thing. The link_graph is where NPLinker computes the scores from to return the genome_to_metabolome matches or vice versa. @CunliangGeng any thoughts?

[ramanujamr]

Yes you're right Justin. The tsv file is coming from this step: npl.to_tsv(link_graph).

[justinjjvanderhooft]

Okay, thanks, we (@liannette and myself) will look into it further.

[ramanujamr]

Hi Justin, any updates on this?

[justinjjvanderhooft]

@liannette have you been able to look at this?

[liannette]

Yes, the families must be coming from the BiG-SCAPE output file. Would be great if you can share the BiG-SCAPE command/parameters you used. Did you maybe use multiple cutoffs?

[ramanujamr]

Hi Annette,
The details commands I used are in the nplinker.txt and run_nplinker.txt files I shared.

I ran npl.load_data() for running BiG-SCAPE through NPLinker.

Here are the parameters used in the toml file:
version = "2"
cutoff = "0.40"
parameters = "--mibig-version 4.0 --include-singletons --gcf-cutoffs 0.40 -c 8"

Thanks for your help!

[liannette]

Hmm, that is strange, then. Is it possible for you to share the data_sqlite.db with us?