|
|
--- |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- protein sequences |
|
|
- biology |
|
|
- bioinformatics |
|
|
task_categories: |
|
|
- text-generation |
|
|
--- |
|
|
The dataset was extracted from antismash-db 4.0 postgresql dump, the corresponding description can be found here: |
|
|
|
|
|
https://antismash-db.secondarymetabolites.org/ |
|
|
|
|
|
If you want to use it, please refer to the original licensing terms and properly cite the authors. |
|
|
|
|
|
Each line in .csv file corresponds to NRPS module, both with the monomer produced by it. |
|
|
Each module has a specific sequence of domains. Module possible structure is described in the literature. |
|
|
To clean the data, I referred to wiki definitions |
|
|
https://en.wikipedia.org/wiki/Nonribosomal_peptide#Modules |
|
|
|
|
|
Each line has the corresponding columns, if the column contains protein sequence, it means that module has the corresponding domain. |
|
|
|
|
|
The monomers are cleaned to keep only aminoacid monomers and corresponding D-aminoacids |
|
|
(it should be defined by the presence of Epimerization domain). |
