Step 1: UniProtKB Processing

Step 1.1: Extract Database Files

tar -xvzf knowledgebase2019_01.tar.gz

Expected directory structure:

knowledgebase2019_01/
├── uniprot_sprot.dat.gz
├── uniprot_trembl.dat.gz
├── docs/
└── reldate.txt

Step 1.2: Create Protein Mapping File

Generate a mapping file containing UniProt protein accessions with their associated metadata.

Expected output: protein_accession-combined.csv
Columns: ID, accession, ncbi_taxonomy_id, pfam_ids, ec_number

# Process Swiss-Prot database
python3 parse_uniprotkb_dat_v2.py \
    -i uniprot_sprot.dat.gz \
    -o protein_accession-sprot.csv \
    -d ','

# Process TrEMBL database
python3 parse_uniprotkb_dat_v2.py \
    -i uniprot_trembl.dat.gz \
    -o protein_accession-trembl.csv \
    -d ','

# Combine both files
bash run_concat.sh

---

Next: Step 2 — UniRef90 Processing