I have been getting good structure predictions for medium and short sequences, of length somewhere around 212 or less.
I had already noticed some larger sequences are rather troublesome. Today I tried running a sequence of length 400 or so, and get a Segmentation Fault error. The previous steps are actually successful, including secondary structure prediction and fragment files generation. I also set my system stack size and max locked memory to unlimited.
Being kind of new to Rosetta, is there something like a maximum sequence length for structure prediction?
Or maybe it's an error in my compilation? System memory allocation perhaps?
Finally, I am using Debian 5.0 64-bit on an Intel Xeon SMP.
Here is the output I get from running AbinitioRelax.linuxgccrelease:
core.init: command: /home/jcsaborio/Rosetta3.0/rosetta3_source/bin/AbinitioRelax.linuxgccrelease -in::file::fasta 1706.fasta -database /home/jcsaborio/Rosetta3.0/rosetta3_database/ -in:file:frag9 aa1706_09_05.200_v1_3 -in:file:frag3 aa1706_03_05.200_v1_3 -out:pdb true -out:nstruct 250 -abinitio:relax true -out:file:silent 1706.out
core.init: 'RNG device' seed mode, using '/dev/urandom', seed=-1019192820 seed_offset=0 real_seed=-1019192820
core.init.random: RandomGenerator:init: Normal mode, seed=-1019192820 RG_type=mt19937
protocols.abinitio.AbrelaxApplication: read fasta sequence: 410 residues
Any help will be greatly appreciated, thanks in advance.