diff --git a/barrOs.py b/barrOs.py
index f9e28751be4441e7139f1e93637f44b3bbb8dc81..bc26114d0b68e61dc104e86d4cb8fbd211135101 100644
--- a/barrOs.py
+++ b/barrOs.py
@@ -9,7 +9,7 @@ number_of_jobs = 1
barros.print_hello()
# Get inputs
-input_files, input_types, input_mode, outf, complexes_only, multimer_only, delete, distance_threshold = barros.get_inputs(sys.argv)
+input_files, input_types, input_mode, outf, complexes_only, multimer_only, delete, distance_threshold, extract_hairpins = barros.get_inputs(sys.argv)
# Check if inputs are correct in case we are not dealing with a 'help' check
@@ -36,7 +36,7 @@ if __name__ == '__main__':
# Prepare all parallel jobs and run the main barrOs method for each pdbid
separated_jobs = barros.chunk_list(pdbIDs, number_of_jobs)
- list_arguments = [i for i in zip(range(number_of_jobs), [input_mode for job in separated_jobs],[complexes_only for job in separated_jobs],[multimer_only for job in separated_jobs],[delete for job in separated_jobs],[distance_threshold for job in separated_jobs], separated_jobs)]
+ list_arguments = [i for i in zip(range(number_of_jobs), [input_mode for job in separated_jobs],[complexes_only for job in separated_jobs],[multimer_only for job in separated_jobs],[delete for job in separated_jobs],[distance_threshold for job in separated_jobs],[extract_hairpins for job in separated_jobs], separated_jobs)]
pool = mp.Pool(number_of_jobs)
data = pool.map(barros.run_barros, list_arguments)
diff --git a/barrOs_library.py b/barrOs_library.py
index cb1066c2af3810781943bcf45d320d0a3687384d..fa967fa9ae407bb5df1454ab58ece1e76c9aeac4 100644
--- a/barrOs_library.py
+++ b/barrOs_library.py
@@ -53,7 +53,7 @@ def print_hello():
print("\n WELCOME TO BARRoS v0 ")
print(" Let me find all barrels in your PDBs ")
print(" ")
- print(" Last change: 06.08.2023 Joana Pereira \n")
+ print(" Last change: 28.02.2023 Joana Pereira \n")
print(" Created at: MPI for Developmental Biology (Protein Evolution) ")
print(" Maintained at: Biozentrum (Protein Structural Bioinformatics) \n")
print("-------------------------------------------------------------------\n")
@@ -62,17 +62,18 @@ def print_help():
print("\nUsage: python3 barrOs.py -in:<type>:<input_file> -out:<mode>\n")
print("Parameters:")
- print(" -in:<type> \tan input string (or list of strings) with one of the following types: {}".format(accepted_input_types))
- print(" \tfor example: -in:pdbID:1LML_A,3XML_A or -in:pdb_file:1LML_A.pdb,3XML_A.pdb")
- print(" \tATENTION: barrOs only works with monomeric proteins, so always provide the target chain")
- print(" -mode: \tdefines the mode of running, i.e. either we want to deal only with membrane proteins, all, or non-membrane proteins")
- print(" \tvalues accepted: {}".format(accepted_modes))
- print(" -outputf: \tdefines the name of the output file")
- print(" \tit is not mandatory. Default: BARRoS_results.csv")
- print(" -nodelete: \tflag to define if files downloaded but without ")
- print(" \tdetected barrels should not be deleted. Default: False")
- print(" -strandist: \tmaximum distance between strands for a contact")
- print(" \tto be considered. Default: 5 Ang")
+ print(" -in:<type> \tan input string (or list of strings) with one of the following types: {}".format(accepted_input_types))
+ print(" \tfor example: -in:pdbID:1LML_A,3XML_A or -in:pdb_file:1LML_A.pdb,3XML_A.pdb")
+ print(" \tATENTION: barrOs only works with monomeric proteins, so always provide the target chain")
+ print(" -mode: \tdefines the mode of running, i.e. either we want to deal only with membrane proteins, all, or non-membrane proteins")
+ print(" \tvalues accepted: {}".format(accepted_modes))
+ print(" -outputf: \tdefines the name of the output file")
+ print(" \tit is not mandatory. Default: BARRoS_results.csv")
+ print(" -nodelete: \tflag to define if files downloaded but without ")
+ print(" \tdetected barrels should not be deleted. Default: False")
+ print(" -strandist: \tmaximum distance between strands for a contact")
+ print(" \tto be considered. Default: 5 Ang")
+ print(" -extract_hairpins: \tflag to extract beta-hairpins, always breaking on the shortest loop. Default: False")
def print_summary(input_files, input_types, input_mode):
@@ -90,6 +91,7 @@ def get_inputs(argv):
complexes_only = False
multimer_only = False
delete = True
+ extract_hairpins = False
input_files = []
input_types = []
input_type = 'nan'
@@ -120,6 +122,8 @@ def get_inputs(argv):
multimer_only = True
elif '-nodelete' in arg:
delete = False
+ elif '-extract_hairpins' in arg:
+ extract_hairpins = True
elif '-strandist' in arg:
distance_threshold = int(arg.split(':')[1])
@@ -133,7 +137,7 @@ def get_inputs(argv):
elif not found_input and found_mode:
input_mode = tmp_mode
- return input_files, input_types, input_mode, output_file, complexes_only, multimer_only, delete, distance_threshold
+ return input_files, input_types, input_mode, output_file, complexes_only, multimer_only, delete, distance_threshold, extract_hairpins
## 1.3. Functions to check if the inputs are correct
@@ -1483,6 +1487,50 @@ def get_barrel_diameter(pdb_file, chainID, min_residue_numb = 3, mode = 'CA', st
else:
return None, None
+def save_hairpins(pdb_sequence, pdb_file, barrel_topology, chains, marker='M', n=2, outfolder = 'fragments'):
+
+ CA_atoms, res_nums = get_CA_coordinates(pdb_file)
+ regular_regions = extract_regular_regions(barrel_topology, pdb_file, marker = marker, add_to_pdb = False, size_threshold = 0)
+
+ register = find_motif_register(regular_regions, n)
+
+ if register is not None:
+ hairpins = [regular_regions[i:i+n] for i in range(register, len(regular_regions), n)]
+
+ outfolder = '{}/{}'.format('/'.join(pdb_file.split('/')[:-1]), outfolder)
+ if not os.path.isdir(outfolder):
+ os.mkdir(outfolder)
+
+ with open(pdb_file, 'r') as inpdb:
+ for line in inpdb:
+ if line.startswith('ATOM') or (line.startswith('HETATM') and 'MSE' in line):
+ resnum = int(line[22:26].strip())
+ chain = line[21:22].strip()
+
+ for i, hairpin in enumerate(hairpins):
+ if len(hairpin) == n:
+ outfile = '{}/{}_fragment{}.pdb'.format(outfolder, pdb_file.split('/')[-1].strip('.pdb'), i)
+
+ start = res_nums[hairpin[0][0]]
+ end = res_nums[hairpin[1][1]]
+
+ if resnum >= start-1 and resnum <= end+1:
+ with open(outfile, 'a+') as outpdb:
+ outpdb.write(line)
+
+
+def find_motif_register(regular_regions, n):
+
+# It finds the register that allows for the shorter linker between units
+
+ starting_motifs = regular_regions[:n+1]
+ linkers = [starting_motifs[i+1][0]-starting_motifs[i][1] for i in range(len(starting_motifs)-1)]
+
+ try:
+ return linkers.index(max(linkers))
+ except:
+ return None
+
## 7. BASIC MATH-RELATED FUNCTIONS
## 7.1. Graphs
@@ -2250,7 +2298,7 @@ def plot_parameter(x_col, y_col, df, saveto, fit_line = False):
def run_barros(arguments, offset = 1, step = 2, local_angle_threshold = 25, max_loop_size = 0):
- job_number, input_mode, complexes_only, multimer_only, delete, distance_threshold, in_queue = arguments
+ job_number, input_mode, complexes_only, multimer_only, delete, distance_threshold, extract_hairpins, in_queue = arguments
# create output files to save the sequences
outfasta = "full_sequences_matched_pdbs_job{}.fasta".format(job_number)
@@ -2466,6 +2514,12 @@ def run_barros(arguments, offset = 1, step = 2, local_angle_threshold = 25, max_
print('... ... There is a barrel in {}_{} but it does not cross the membrane\n'.format(pdbID, chainID))
if delete:
os.system("rm {}*".format(pdb_file[:-4]))
+
+ if extract_hairpins and os.path.isfile(barrel_struct):
+ save_hairpins(pdb_sequence, pdb_file, barrel_topology, chains=chains, outfolder='hairpins')
+
+
+
else:
print('... ... Not able to detect barrel topology for {}_{}\n'.format(pdbID, chainID))
if delete:
@@ -2477,6 +2531,10 @@ def run_barros(arguments, offset = 1, step = 2, local_angle_threshold = 25, max_
outbb.write('>NaN_TM_BARREL_{}_{}_{}\n'.format(protein_type, pdbID, chainID))
outbb.write('{}\n'.format(pdb_sequence))
+ if extract_hairpins:
+ save_hairpins(pdb_sequence, pdb_file, seqstruct, marker='E', chains=chains, outfolder='hairpins')
+
+
else:
print(" ... ... pdbID '{}_{}' impossible to get".format(pdbID, chainID))