A strange mistake：Error: msa2lddt died

[tzjsdss]

Got databases
Got substitution matrices
Initialised MSAs, Sequence objects
Set up tiny substitution matrices
Performing initial all vs all alignments
Generating guide tree
Optimising merge order
Writing guide tree to: result.nw
Begin progressive alignment
Time for merging to result_aa.fa: 0h 0m 0s 153ms
Time for merging to result_3di.fa: 0h 0m 0s 143ms
Time for processing: 1h 27m 39s 330ms
result.html exists and will be overwritten
msa2lddtreport tmpFolder/11829316448293657361/structures result_aa.fa result.html --guide-tree result.nw --pair-threshold 0 --threads 12 -v 3 --report-command '--match-ratio 0.51 --filter-msa 1 --gap-open aa:10,nucl:10 --gap-extend aa:1,nucl:1 --report-mode 1 ' --report-paths 1

Invalid database read for id=18446744073709551615, database index=tmpFolder/11829316448293657361/structures.lookup
getLookupKey: local id (18446744073709551615) >= db size (4101)
Error: msa2lddt died

The above is my error log, and I cannot understand how this error is related to the quantity of 4101. My protein quantity is 4099, and I have tried my other protein libraries (also over 4000) that can run and succeed, but my protein library cannot succeed. In fact, after running it, 3di, aa, and nw files were generated (but not json and html files), but I am not sure if this error has affected my results. Please help me explain and solve it. Thank you

Here are my commands
foldmason easy-msa msaah result tmpFolder --precluster

[milot-mirdita]

Which foldmason release/git commit is this?

[tzjsdss]

foldmason-linux-avx2.tar.gz 14-Jan-2025 08:51 54240479

this

[milot-mirdita]

What version does foldmason version report?

Could you try without --precluster? I think we might have some issues with this mode currently.

[tzjsdss]

Yes, I tried and this command won't produce any errors, but I still haven't received JSON or HTML. Actually, I just want to cluster the proteins, and what I most want is a list with scores for each protein a and its similar protein b