comparative-genomics-project/workflow/scripts/protein_lengths.awk

#!/usr/bin/env -S awk -f
# Associate a isoform id
# to the length of the sequence
# and the associated gene id

BEGIN {
    sequence_length=0
    isoform_id=""
    gene_id=""

    OFS="\t"
}

/^>/ {
    if (isoform_id != "") {
        print isoform_id, sequence_length, gene_id
    }
    isoform_id = $1
    gene_id = $4
    gsub(">", "", isoform_id)
    gsub("gene:", "", gene_id)
    sequence_length = 0
}

/^[^>]/ {
    sequence_length += length($0)
}

END {
    if (isoform_id != "") {
        print isoform_id, sequence_length, gene_id
    }
}