Hi,

I am using forking to align several samples in parallel using Rsubread with mclapply. The scalability of this approach depends on the available memory needed to load the index in each forked process, but I think this problem could be avoided by loading the index in shared memory for all the child processes.

Is this currently possible for Rsubread? I have read the documentation and the Google group, but I didn't find this issue being discussed.

Thanks,

Nicolas

No Rsubread does not support sample-level parallelization. It supports read-level parallelization which should achieve at least the same speed-up as the sample-level parallelization. The nthreads parameter in align and subjunc functions allows you to run multithreaded read alignment.