Error, 'object too long to compress'

Question

[felix_langschied]

Dear OMA-Team,

I have run the all-vs-all step in the proteomes of 10 plants, and as far as I can tell, all jobs finished sucessfully. However, when I try to run the clustering step (stage 2), I receive the Error, 'object too long to compress'.

I have run the oma-compact command in the hope to solve the issue but the Error persists. Does anyone know what the source of the Error can be?

Thanks a ton in advance!

Answer 1

[Kevin Blighe]

We are not the OMA team here - did you copy-paste your question from somewhere else to here?

In any case...

The error 'object too long to compress' indicates that the data generated from the all-versus-all step exceeds the size limit for compression during the clustering step. This can occur when the proteomes produce an excessively large number of matches.

Run the oma-cleanup command to remove incomplete or unnecessary files from the Cache/AllAll directory. After that, run the oma-compact command again to summarize the results files.

Ensure that your system uses 64-bit Python and has at least 32 gigabytes of random access memory available for processing 10 plant proteomes. If one or more proteomes contain many long proteins or redundant isoforms, filter the input fasta files to include only canonical proteins before re-running the all-versus-all step.

Kevin