Key ERROR in [filter_gtf_by_txid.err]

[Marison-1]

Hi, thank you for share this nice tools! I met a error when I using galba `Traceback (most recent call last): File "/public-supool/home/marison/software/Augustus-master/scripts/filter_gtf_by_txid.py", line 49, in gene_data[gene_id][transcript_id].append(row)

KeyError: 'g27699.t3'`

and this is command `/public-supool/home/marison/software/GALBA/scripts/galba.pl --gff3 --threads 20 --prg=miniprot --SCORER_PATH=/public-supool/home/marison/software/miniprot-boundary-scorer/ --MINIPROTHINT_PATH=/public-supool/home/marison/software/miniprothint/ --CDBTOOLS_PATH=/public-supool/home/marison/micromamba/envs/braker3/bin/ --genome=/public-dss/share/gap_lab/MaYuSheng/Medicago/Genome/Assembly/ragtag.scaffold.fasta --AUGUSTUS_CONFIG_PATH=/public-supool/home/marison/software/Augustus-master/CONFIG/ --AUGUSTUS_BIN_PATH=/public-supool/home/marison/software/Augustus-master/bin/ AUGUSTUS_SCRIPTS_PATH=/public-supool/home/marison/software/Augustus-master/scripts/ --species=TLMX --prot_seq=/public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/ProteinSeq/EvidenceProtein2.fasta `

I wonder if there are some details that I hadn't noticed or it's just a small bug in [filter_gtf_by_txid.py]? thank you very much!

[KatharinaHoff]

it would help to have the full call of filter_gtf_by_txid.py (it is in the GALBA.log file), and the related input files that lead to the error message.

[Marison-1]

it would help to have the full call of filter_gtf_by_txid.py (it is in the GALBA.log file), and the related input files that lead to the error message.

Hi, thank you very much for reply! This is the last report of [galba.log] file which lead to the error # Thu Mar 28 04:27:26 2024: deleting transcripts with CDS features on opposite strands and with overlapping CDS features... /public-supool/home/marison/micromamba/envs/galba/bin/python3 /public-supool/home/marison/software/GALBA/scripts/gtf_sanity_check.py -f /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/augustus.tmp.gtf -o /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/bad2.lst > /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/gtf_sanity_check.log 2> /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/errors/gtf_sanity_check.err /public-supool/home/marison/micromamba/envs/galba/bin/python3 /public-supool/home/marison/software/Augustus-master/scripts/filter_gtf_by_txid.py -g /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/augustus.tmp.gtf -l /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/bad2.lst -o /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/augustus.hints.gtf > /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/filter_gtf_by_txid.log 2> /public-dss/share/gap_lab/MaYuSheng/Medicago/GeneAnno/galba/TLMX_Modified/GALBA/errors/filter_gtf_by_txid.err, and this is the input data file screen shot where [KeyError: 'g27699.t3'] happend:

furthermore,this is the script named "filter_gtf_by_txid.py" in augustus scripts directory `#!/usr/bin/env python3

author = "Katharina J. Hoff with a little help of ChatGPT" copyright = "Copyright 2022-2023. All rights reserved." credits = "Katharina Hoff" version = "1.0.1" email = "katharina.hoff@uni-greifswald.de" status = "development"

Import the necessary libraries

import csv import re import argparse

parser = argparse.ArgumentParser(description='Remove transcripts form a gtf file') parser.add_argument('-g', '--gtf', required=True, type=str, help="Name of (sorted) gtf file") parser.add_argument('-l', '--exclude_lst', required=True, type=str, help="Text file with transcript IDs") parser.add_argument('-o', '--output_file', required=True, type=str, help="Output file in gtf format") args = parser.parse_args()

Initialize a set to store the IDs of the transcripts to remove

transcripts_to_remove = set()

Read the transcript IDs from the file and add them to the set

with open(args.exclude_lst, "r") as transcript_ids_file: for line in transcript_ids_file: transcripts_to_remove.add(line.strip())

Initialize the dictionary to store the gene and transcript data

gene_data = {}

Open the GTF file and use the csv reader to read through the rows

with open(args.gtf, "r") as gtf_file: reader = csv.reader(gtf_file, delimiter="\t") for row in reader: if row[2] == "CDS" or row[2]=="exon" or row[2]=="intron" or row[2]=="start_codon" or row[2]=="stop_codon":

Extract the transcript ID from the row

  transcript_id = row[8].split("transcript_id ")[1].split(";")[0].strip('"')
  gene_id = row[8].split("gene_id ")[1].split(";")[0].strip('"')
elif row[2]=="transcript":
  transcript_id = row[8].strip()
elif row[2]=="gene":
  gene_id = row[8].strip()
if row[2]=="gene":
  gene_data[gene_id] = {"gene_line": row}
elif row[2]=="transcript":
  gene_data[gene_id][transcript_id]=[row]
else:
  gene_data[gene_id][transcript_id].append(row)

for tx in transcripts_to_remove: match = re.match(r'([^"]*g\d+).t\d+', tx) gid = match.group(1) if gid in gene_data: if tx in gene_data[gid]: del gene_data[gid][tx] if gid in gene_data: if len(gene_data[gid]) == 1: del gene_data[gid]

Open the GTF file and use the csv reader to read through the rows

with open(args.output_file, "w") as safe: for key, value in gene_data.items(): safe.write("\t".join(value['gene_line']) + "\n") del value['gene_line'] for innerkey in value.keys(): for ele in value[innerkey]: safe.write("\t".join(ele) + "\n")

`

the error report indicate that 49 line of above script has a KeyError in my data,I checked the them again and foud that the directory {"gene_data[gene_id]"} needs a key["transcript_id"] which require the column[2] is 'transcript' (elif row[2]=="transcript":), but in my data,[g27699.t3] seems dosen't have a transcript type, which may lead to the error.

If this kind of situation is common, I hope it will be helpful for script modification or it's just a data problem.

Thanks for your reply again!