Alternate coding table functionality is broken

[usmanenam]

Hello,

I saw that in v1.4.0 you fixed an issue where when you pass in a different coding table, the final gbk ends up with translations from the standard coding table. I am experiencing that issue now many years later. If I pass in an alternate code (in single mode, and whether the gene_caller is specified or not), I end up with genes that are with coding table 11. If I pass in prodigal as the gene caller, I still get genes that are in coding table 11 even though they say the new coding table number.

[gbouras13]

Hi @usmanenam - can you share an example?

George

[usmanenam]

For example, this command
pharokka.py -i ./phage.fa -o ./output_tr6/ -d ../../pharokka/pharokkadb/ -t 16 -f -c 6 -g prodigal

results in a gbk file where the correct qualifier is given (/transl_table=6) but the actual translations are from translation table 11. This is confirmed from the prodigal.ffn file where there are genes ending in TAA and TAG (these should be read as Q instead of *).

When prodigal-gv (or no gene caller) is specified the gbk file /transl_table=11 even when the -c 6 flag is used and it is read with translation table 11.

Is that not happening on your end with any random sequence file?

[gbouras13]

Can you try:

pyrodigal -g 6 -i ./phage.fa

And let me know if that differs from the pharokka output?