Started running InterProscan locally. Looks like the process isn't progressing well. Can you look at the report?

Question

[Vijith]

Performing interproscan (v.5.69-101.0) locally with Docker container. My fasta file consists of ~5 lakh sequences. The process looks like it was stuck at 14%. The server machine has 8 cores and 245GB memory. Herewith, I have provided the report hitherto. Can someone provide any valuable input?

report: https://docs.google.com/document/d/1RH8SHBXB1UJ-b6zu6ePkLnFDcf4DcUsTRYVZQKsXBvU/edit?usp=sharing

[dthorbur]

As in 500,000 sequences? If it's getting stuck, it might be better to split the input into smaller chunks and loop through them instead. At least if there are a set of sequences that are causing interproscan to hang, you'll be able to identify which more easily.

[colindaven]

Its always good practice to check if a process is "stuck" by looking at

• RAM - htop, top
• CPU usage - are the cpus still occupied ? htop, bottom etc.
• hard disk or network activity - glances, bottom etc
• log files and working directories - are outputs being written, increasing in size. When was the last write ? ls -lh, etc.

Answer 1

[Olivier]

You can try to process sequences in chunks. The image https://hub.docker.com/r/agbase/interproscan, which is a wrapper around the official interproscan image, should do the trick (I've not tested it though). But I do have tested to process sequences in chunks manually, and it solved lots of issues