Block table may be destroyed

[cfz1998]

Hi! I get a error!

Gd:0 No:0 My:4096 MS:0 Mb:21588 Tw:0 Tl:123750096 0.00 100.00 0.00
: Block table may be destroyed !

What's means about this! When i used braker2 pipeline.

spaln-master/bin/spaln -Q7 -O0 genomeCq00.fa uniprot_plants_addstop.fa
Gd:0 No:0 My:4096 MS:0 Mb:21588 Tw:0 Tl:123750096 0.00 100.00 0.00
: Block table may be destroyed !

[ogotoh]

This message is displayed when you have not yet formatted your genomic sequence. Follow the instruction of this GitHub README.md for formatting. Typically, you may run spaln in the directory where your genomic sequence, genome.fa in FASTA format, resides:
$ spaln -W -K[D|P] genome.fa
According to your query sequence type (DNA/RNA or protein), you may choose ‘D’ or ‘P’ of the -K option.

[crysclitheroe]

Good day,

I have a similar problem while spaln is running automatically in braker2; It seems that at the braker2 step:

OUTPUT OF makblk.pl STARTING

Thu Jul 21 11:24:17 2022: run spaln for target '/flash/braker2/tmp_spaln/genome.fa' and query '/flash/braker2/tmp_spaln/Cockroach_Termite_InsectaTranscripts_fromUniprot_addstop.fa'

/home/ProtHint/dependencies/spaln -Q7 -O0 /flash/braker2/tmp_spaln/genome.fa /flash/braker2/tmp_spaln/Cockroach_Termite_InsectaTranscripts_fromUniprot_addstop.fa > /flash/braker2/align_spaln/spaln.aln 2>/flash/braker2/align_spaln/spaln.stderr

Spaln seems to be formatting the target genome.fa incorrectly, with -KP option, as opposed to -KD option:
startAlign.stderr.txt

Which creates the Block table may be destroyed error in the spaln.stderr file.

Is this correct or is something configured incorrectly?

[ogotoh]

Thank you for your report, according to which I can figure out the cause of the error.

Looking into the error message you provided, the problem seems to reside in the startAlign.pl script. The command

…/spaln -Q7 -O0 /flash/braker2/tmp_spaln/genome.fa …Transcripts….fa

lacks -d option which is required to access to the formatted genomic sequences. The command would have to be

…/spaln -Q7 -O0 -d genome …Transcripts….fa

assuming that the “seqdb” directory is set to /flash/braker2/tmp_spaln, e.g. by
export ALN_DBS=/flash/braker2/tmp_spaln

Two additional comments:

1. The startAlign.pl script could work in the present form provided that one-time formatting facility (see the document) works adequately. However, this option is formally supported only for protein databases with -KA option. For -KD and -KP options, its behavior has not been well tested, and probably fails at a high probability. Note also that in this setting, formatting is redundantly done.
2. You need not format transcript sequences.

Osamu,