updated with temporay crystfel version

This commit is contained in:
Beale John Henry
2024-01-31 11:06:17 +01:00
parent 201ead6441
commit 0464e77288
4 changed files with 38 additions and 23 deletions

View File

@@ -150,7 +150,8 @@ def write_crystfel_run( proc_dir, name, chunk, chunk_lst_file, geom_file, cell_f
run_sh = open( cryst_run_file, "w" )
run_sh.write( "#!/bin/sh\n\n" )
run_sh.write( "module purge\n" )
run_sh.write( "module load crystfel/0.10.2\n" )
run_sh.write( "module use MX unstable\n" )
run_sh.write( "module load crystfel/0.10.2-rhel8\n" )
run_sh.write( "indexamajig -i {0} \\\n".format( chunk_lst_file ) )
run_sh.write( " --output={0} \\\n".format( stream_file ) )
run_sh.write( " --geometry={0} \\\n".format( geom_file ) )

View File

@@ -5,7 +5,8 @@
"""
# aim
to process a batch of data very fast by splitting it into a number of chunks and submitting
these jobs separately to the cluster
these jobs separately to the cluster - but now all with the ability to change crystfel parameters
from the command line
# usage
python crystfel_split.py -l <path-to-list-file>
@@ -13,7 +14,6 @@ python crystfel_split.py -l <path-to-list-file>
-g <path-to-geom-file>
-c <path-to-cell-file>
-n <job-name> -default split
# note -p progress bar (True/False) -default True
-t crystfel threshold -default 10
-s crystfel min-snr -default 5
-i crystfel int-radius -default 3,5,9
@@ -54,7 +54,7 @@ def h5_split( lst, chunk_size ):
def write_crystfel_run( proc_dir, name, chunk, chunk_lst_file,
geom_file, cell_file, threshold, min_snr,
int_rad, multi, retry, min_pix ):
int_rad, multi, retry, min_pix, bg_rad, min_res ):
# stream file name
stream_file = "{0}_{1}.stream".format( name, chunk )
@@ -65,20 +65,27 @@ def write_crystfel_run( proc_dir, name, chunk, chunk_lst_file,
run_sh = open( cryst_run_file, "w" )
run_sh.write( "#!/bin/sh\n\n" )
run_sh.write( "module purge\n" )
run_sh.write( "module load crystfel/0.10.2\n" )
run_sh.write( "module use MX unstable\n" )
run_sh.write( "module load crystfel/0.10.2-rhel8\n" )
run_sh.write( "indexamajig -i {0} \\\n".format( chunk_lst_file ) )
run_sh.write( " --output={0} \\\n".format( stream_file ) )
run_sh.write( " --geometry={0} \\\n".format( geom_file ) )
run_sh.write( " --pdb={0} \\\n".format( cell_file ) )
run_sh.write( " --indexing=xgandalf-latt-cell --peaks=peakfinder8 \\\n" )
run_sh.write( " --indexing=xgandalf-latt-cell \\\n" )
run_sh.write( " --peaks=peakfinder8 \\\n" )
run_sh.write( " --integration=rings-grad \\\n" )
run_sh.write( " --tolerance=10.0,10.0,10.0,2,3,2 \\\n" )
run_sh.write( " --threshold={0} \\\n".format( threshold ) )
run_sh.write( " --min-snr={0} \\\n".format( min_snr ) )
run_sh.write( " --int-radius={0},{1},{2} \\\n".format( int_rad[0], int_rad[1], int_rad[2] ) )
run_sh.write( " -j 32 --max-res=3000\\\n" )
run_sh.write( " -j 32 \\\n" )
run_sh.write( " --{0} \\\n".format( multi ) )
run_sh.write( " --check-peaks \\\n" )
run_sh.write( " --{0} \\\n".format( retry ) )
run_sh.write( " --min-pix-count={0}\\\n".format( min_pix ) )
run_sh.write( " --min-res=85\\\n" )
run_sh.write( " --max-res=3000 \\\n" )
run_sh.write( " --min-pix-count={0} \\\n".format( min_pix ) )
run_sh.write( " --local-bg-radius={0} \\\n".format( bg_rad ) )
run_sh.write( " --min-res={0}".format( min_res ) )
run_sh.close()
# make file executable
@@ -175,8 +182,6 @@ def run_splits( cwd, name, lst, chunk_size, geom_file,
print( "DONE" )
# include progress bar if required
if progress==True:
wait_for_jobs(submitted_job_ids, chunk)
print("slurm processing done")
@@ -237,13 +242,6 @@ if __name__ == "__main__":
type=str,
default="split"
)
parser.add_argument(
"-p",
"--progress",
help="gives you the option of also having a progress bar",
type=bool,
default=True
)
parser.add_argument(
"-t",
"--threshold",
@@ -286,6 +284,20 @@ if __name__ == "__main__":
type=int,
default=2
)
parser.add_argument(
"-b",
"--bg_rad",
help="crystfel background radius flag, radius (in pixels) used for the estimation of the local background",
type=int,
default=2
)
parser.add_argument(
"-q",
"--min-res",
help="m",
type=int,
default=2
)
args = parser.parse_args()
# run geom converter
cwd = os.getcwd()
@@ -298,6 +310,6 @@ if __name__ == "__main__":
else:
retry = "no-retry"
run_splits( cwd, args.job_name, args.lst_file, args.chunk_size,
args.geom_file, args.cell_file, args.progress,
args.geom_file, args.cell_file,
args.threshold, args.min_snr, args.int_radius,
multi, retry, args.min_pix )

View File

@@ -71,7 +71,8 @@ def run_partialator( proc_dir, name, stream, pointgroup, model, iterations, cell
part_sh = open( part_run_file, "w" )
part_sh.write( "#!/bin/sh\n\n" )
part_sh.write( "module purge\n" )
part_sh.write( "module load crystfel/0.10.2\n" )
part_sh.write( "module use MX unstable\n" )
part_sh.write( "module load crystfel/0.10.2-rhel8\n" )
part_sh.write( "partialator -i {0} \\\n".format( stream ) )
part_sh.write( " -o merged_{0}.hkl \\\n".format( name ) )
part_sh.write( " -y {0} \\\n".format( pointgroup ) )

View File

@@ -73,7 +73,8 @@ def run_partialator( proc_dir, stream, pointgroup, model, iterations, push_res,
part_sh = open( part_run_file, "w" )
part_sh.write( "#!/bin/sh\n\n" )
part_sh.write( "module purge\n" )
part_sh.write( "module load crystfel/0.10.2\n" )
part_sh.write( "module use MX unstable\n" )
part_sh.write( "module load crystfel/0.10.2-rhel8\n" )
part_sh.write( "partialator -i {0} \\\n".format( stream ) )
part_sh.write( " -o push-res_{0}.hkl \\\n".format( push_res ) )
part_sh.write( " -y {0} \\\n".format( pointgroup ) )