def core(self, seq: str) -> None:
self.core_analysis_monoisotopic = utils.RobustProteinAnalysis(
seq, monoisotopic=True)
self.core_analysis = utils.RobustProteinAnalysis(seq,
monoisotopic=False)
self._core = seq
self._calculate_mw()
def core(self, seq: str) -> None:
assert isinstance(seq, str)
assert seq
self._core = seq
self._weight = -1.
self._monoisotopic_weight = -1.
self.core_analysis_monoisotopic = utils.RobustProteinAnalysis(seq, monoisotopic=True)
self.core_analysis = utils.RobustProteinAnalysis(seq, monoisotopic=False)
def test_molecular_weight_ignore(self):
""" Test RobustProteinAnalysis.molecular_weight() calculates
correct weight when ignoring invalids"""
rpa = utils.RobustProteinAnalysis("MAGICXHAT")
self.assertEqual(802.9621, rpa.molecular_weight()) # default is True
rpa = utils.RobustProteinAnalysis("MAGICXHAT", ignore_invalid=True)
self.assertEqual(802.9621, rpa.molecular_weight())
def test_init(self):
"""Test RobustProteinAnalysis initialisation"""
rpa = utils.RobustProteinAnalysis("MAGICHAT", ignore_invalid=True)
self.assertIsInstance(rpa, utils.RobustProteinAnalysis)
rpa = utils.RobustProteinAnalysis("MAGICHAT", ignore_invalid=False)
self.assertIsInstance(rpa, utils.RobustProteinAnalysis)
for bad_invalid in ["none", None, 3, []]:
with self.assertRaises(TypeError):
utils.RobustProteinAnalysis("MAGICHAT",
ignore_invalid=bad_invalid)
def cut_weight(self) -> float:
""" Determines the weight of the core peptide without tail """
if not self.core:
raise ValueError("Cannot calculate cut weights without a core")
if not self.c_cut:
return self.molecular_weight
return self._calculate_weight(utils.RobustProteinAnalysis(self.core[:-len(self.c_cut)], monoisotopic=False))
def _calculate_number_of_bridges(self)-> int:
"""
Predict the lassopeptide number of disulfide bridges
"""
aas = utils.RobustProteinAnalysis(self.core, monoisotopic=True).count_amino_acids()
if aas['C'] >= 4:
return 2
elif aas['C'] >= 2:
return 1
return 0
def number_bridges(self):
"""
Predict the lassopeptide number of disulfide bridges
"""
aas = utils.RobustProteinAnalysis(
self.core, monoisotopic=True).count_amino_acids()
if aas['C'] >= 4:
self._num_bridges = 2
elif aas['C'] >= 2:
self._num_bridges = 1
return self._num_bridges
def test_molecular_weight_average(self):
""" Test RobustProteinAnalysis.molecular_weight() calculates
correct weight when not ignoring invalids
"""
rpa = utils.RobustProteinAnalysis("MAGICXHAT", ignore_invalid=False)
self.assertEqual(912.9621, rpa.molecular_weight())
def test_uppercase(self):
"""Test RobustProteinAnalysis converts passed sequence to upper case"""
rpa = utils.RobustProteinAnalysis("Magichat")
assert rpa.original_sequence == "MAGICHAT"
assert rpa.sequence == "MAGICHAT"
def acquire_rodeo_heuristics(record: secmet.Record, query: secmet.CDSFeature,
leader: str, core: str,
domains: List[str]) -> Tuple[int, List[float]]:
""" Calculate heuristic scores for RODEO
Arguments:
record: the record instance to analyse
query: the feature being checked
leader: the sequence of the peptide leader
core: the sequence of the peptide core
domains: the domains found within CDS features of the cluster
Returns:
a tuple of
the RODEO score, and
a list of floats for use in the RODEO SVM
"""
tabs = [] # type: List[float]
score = 0
precursor = leader + core
# Leader peptide contains FxLD motif
if re.search('F.LD', leader):
score += 2
tabs.append(1)
else:
tabs.append(0)
# Core residue position of Sx4C motif
match = re.search('S....C', core)
if match:
tabs.append(match.span()[0])
else:
tabs.append(0)
# Core residue position of Tx4C motif
match = re.search('T....C', core)
if match:
tabs.append(match.span()[0])
else:
tabs.append(0)
# Core residue position of Sx5C motif
match = re.search('S.....C', core)
if match:
tabs.append(match.span()[0])
else:
tabs.append(0)
# Core residue position of Tx5C motif
match = re.search('T.....C', core)
if match:
tabs.append(match.span()[0])
else:
tabs.append(0)
# Precursor is within 500 nt?
hmmer_profiles = ['LANC_like', 'Lant_dehyd_C']
distance = utils.distance_to_pfam(record, query, hmmer_profiles)
if distance < 500:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Cluster contains LanB dehydratase domain (PF04738)
if "Lant_dehyd_C" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Cluster contains Lan C cyclase domain (PF05147)
if "LANC_like" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Cluster LACKS LanB dehydratase domain (PF04738)
if "Lant_dehyd_C" not in domains:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Cluster LACKS Lan C cyclase domain (PF05147)
if "LANC_like" not in domains:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Cluster contains LanB dehydratase elimination C-terminal domain (PF14028)
if "PF14028" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Cluster contains S8 peptidase subtilase (PF00082)
if "Peptidase_S8" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Cluster contains C39 peptidase (PF03412)
if "Peptidase_C39" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Cluster contains ABC transporter (PF00005)
if "PF00005" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Cluster contains YcaO-like protein (PF02624)
if "YcaO" in domains:
score -= 4
tabs.append(1)
else:
tabs.append(0)
# Cluster contains ThiF-like protein (PF00899)
if "ThiF" in domains:
score -= 4
tabs.append(1)
else:
tabs.append(0)
# Cluster contains PF02052 (Gallidermin)
if set(domains).intersection(
{"Gallidermin", "mature_a", "mature_b", "matura_ab"}):
tabs.append(1)
else:
tabs.append(0)
# Cluster contains PF8130
if "Antimicr18" in domains:
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide mass < 4000 Da
precursor_analysis = utils.RobustProteinAnalysis(precursor,
monoisotopic=True,
ignore_invalid=True)
if precursor_analysis.molecular_weight() < 4000:
score -= 3
tabs.append(1)
else:
tabs.append(0)
# Core peptide mass < 2000 Da
core_analysis = utils.RobustProteinAnalysis(core,
monoisotopic=True,
ignore_invalid=True)
if core_analysis.molecular_weight() < 2000:
score -= 3
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide pHMMs below:
precursor_hit = False
# Precursor peptide hits gallidermin superfamily (cl03420) HMM
if cds_has_domains(query, {"TIGR03731", "Gallidermin"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide hits lantibio_gallid (TIGR03731) HMM
if cds_has_domains(query, {"TIGR03731"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide hits lanti_SCO0268 superfamily (cl22812) HMM
if cds_has_domains(query, {"TIGR04451", "strep_PEQAXS"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide hits LD_lanti_pre (TIGR04363) HMM
if cds_has_domains(query, {"LD_lanti_pre"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide hits Antimicrobial18 (cl06940) HMM
if cds_has_domains(query, {"Antimicr18"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# Precursor peptide hits gallidermin (PF02052) HMM
if cds_has_domains(query,
{"Gallidermin", "mature_a", "mature_ab", "mature_b"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
# precursor peptide hits Antimicrobial18 (PF08130) HMM
if cds_has_domains(query, {"Antimicr18"}):
precursor_hit = True
tabs.append(1)
else:
tabs.append(0)
if precursor_hit:
score += 3
# Precursor peptide mass (unmodified)
precursor_analysis = utils.RobustProteinAnalysis(precursor,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(precursor_analysis.molecular_weight()))
# Unmodified leader peptide mass
leader_analysis = utils.RobustProteinAnalysis(leader,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(leader_analysis.molecular_weight()))
# Unmodified core peptide mass
core_analysis = utils.RobustProteinAnalysis(core,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(core_analysis.molecular_weight()))
# Length of leader peptide
tabs.append(len(leader))
# Length of core peptide
tabs.append(len(core))
# Length of precursor peptide
tabs.append(len(precursor))
# Ratio of length of leader peptide / length of core peptide
tabs.append(float(len(leader) / float(len(core))))
# Core peptide ≥ 35 residues
if len(core) >= 35:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Core peptide contains CC motif (not in last 3 residues)
if 'CC' in core[:-3]:
score -= 3
tabs.append(1)
else:
tabs.append(0)
# Leader peptide has > 4 negatively charge motifs
if sum([leader.count(aa) for aa in "DE"]) > 4:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader peptide has net negative charge
charge_dict = {"E": -1, "D": -1, "K": 1, "R": 1}
if sum([charge_dict[aa] for aa in leader if aa in charge_dict]) < 0:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader residue position of FxLD motif
match = re.search('F.LD', leader)
if match:
tabs.append(match.span()[0])
else:
tabs.append(0)
# Core peptide contains C-terminal CC (within last 3 residues)
if 'CC' in core[-3:]:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Core peptide contains DGCGxTC / SFNS / SxxLC / CTxGC / TPGC / SFNSxC motifs
motifs = (('DGCG.TC', 2), ('SFNS', 2), ('S..LC', 2), ('CT.GC', 1),
('TPGC', 1), ('SFNS.C', 1))
for motif, motif_score in motifs:
if re.search(motif, core):
score += motif_score
tabs.append(1)
else:
tabs.append(0)
# Core peptide contains < 2 or < 3 Cys
if core.count("C") < 2:
score -= 6
tabs += [1, 1]
elif core.count("C") < 3:
score -= 3
tabs += [1, 0]
else:
tabs += [0, 0]
# No Cys/Ser/Thr in core peptide
for amino, penalty in [("C", -10), ("S", -4), ("T", -4)]:
if amino not in core:
score += penalty
tabs.append(1)
else:
tabs.append(0)
# Lanthionine regex maximum ring number > 4
numrings, profile = lanscout(core)
if numrings > 4:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Lanthionine regex maximum ring number < 3
if numrings < 3:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Lanthionine regex 4-membered ring/5-membered ring/6-membered ring/7-membered ring/8-membered ring
scores = [2, 2, 2, 2, 1]
scorepos = 0
for ringsize in profile[:2]:
if ringsize not in [0, 1, 2]:
score += scores[scorepos]
tabs.append(1)
else:
tabs.append(0)
scorepos += 1
for ringsize in profile[2:]:
if ringsize != 0:
score += scores[scorepos]
tabs.append(1)
else:
tabs.append(0)
scorepos += 1
return score, tabs
def acquire_rodeo_heuristics(
cluster: secmet.Protocluster, query: secmet.CDSFeature, leader: str,
core: str, domains: Dict[str,
int]) -> Tuple[int, List[float], List[int]]:
"""Calculate heuristic scores for RODEO"""
tabs = []
score = 0
precursor = leader + core
# Calcd. precursor peptide mass (Da)
precursor_analysis = utils.RobustProteinAnalysis(precursor,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(precursor_analysis.molecular_weight()))
# Calcd. leader peptide mass (Da)
leader_analysis = utils.RobustProteinAnalysis(leader,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(leader_analysis.molecular_weight()))
# Calcd. core peptide mass (Da)
core_analysis = utils.RobustProteinAnalysis(core,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(core_analysis.molecular_weight()))
# Distance to any biosynthetic protein (E, B, C)
hmmer_profiles = ['PF04055']
distance = utils.distance_to_pfam(cluster.parent_record, query,
hmmer_profiles)
tabs.append(distance)
# rSAM within 500 nt?
if utils.distance_to_pfam(cluster.parent_record, query, ['PF04055']) < 500:
score += 1
tabs.append(1)
else:
tabs.append(0)
# rSAM within 150 nt?
if utils.distance_to_pfam(cluster.parent_record, query, ['PF04055']) < 150:
score += 1
tabs.append(1)
else:
tabs.append(0)
# rSAM further than 1000 nt?
if utils.distance_to_pfam(cluster.parent_record, query, ['PF04055']) == -1 or \
utils.distance_to_pfam(cluster.parent_record, query, ['PF04055']) > 10000:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Ratio of N-term to 1st Cys 0.25<x<0.60; Ratio of N-term to 1st Cys <0.25 or >0.60
if "C" not in precursor:
score -= 2
tabs += [0, 1]
elif 0.25 <= precursor.find("C") / len(precursor) <= 0.60:
score += 2
tabs += [1, 0]
else:
score -= 2
tabs += [0, 1]
# Three or more Cys; Less than 3 Cys
if precursor.count("C") >= 3:
score += 4
tabs += [1, 0]
else:
score -= 4
tabs += [0, 1]
# CxC/CxxC/CxxxC/CxxxxxC; # CC/CCC
motifs = (('C.{5}C', 2), ('C.{3}C', 1), ('C.{2}C', 1), ('C.{1}C', 1),
('CC', -2), ('CCC', -2))
for motif in motifs:
if re.search(motif[0], core):
score += motif[1]
tabs.append(1)
else:
tabs.append(0)
# No Cys in last 1/4th?
quarter_length = -len(precursor) // 4
if "C" not in precursor[quarter_length:]:
score += 1
tabs.append(1)
else:
score -= 1
tabs.append(0)
# 2 Cys in first 2/3rds of precursor, 1 Cys in last 1/3rd of precursor
two_thirds = 2 * len(precursor) // 3
if precursor[:two_thirds].count("C") == 2 and precursor[two_thirds:].count(
"C") == 1:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Peptide matches SboA hmm
if cds_has_domains(query, {"Subtilosin_A"}):
score += 3
tabs.append(1)
else:
tabs.append(0)
# Peptide matches SkfA hmm
if cds_has_domains(query, {"TIGR04404"}):
score += 3
tabs.append(1)
else:
tabs.append(0)
# Peptide matches SCIFF hmm
if cds_has_domains(query, {"TIGR03973"}):
score += 2
tabs.append(1)
else:
tabs.append(0)
# cluster has PqqD/RRE (PF05402)
if "PF05402" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has SPASM domain (PF13186)
if "PF13186" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# PF04055 (rSAM) domain start > 80
runresults = subprocessing.run_hmmsearch(
path.get_full_path(__file__, "data", "PF04055.hmm"),
fasta.get_fasta_from_features(cluster.cds_children))
max_start = 0
hitstarts = []
hitends = []
for runresult in runresults:
# Store result if it is above cut-off
for hsp in runresult.hsps:
if hsp.bitscore > 40:
hitstarts.append(hsp.hit_start)
max_start = max(hsp.hit_start, max_start)
hitends.append(hsp.hit_end)
if hitstarts and max_start > 80:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has peptidase
peptidase_domains = [
"Peptidase_M16_C", "Peptidase_S8", "Peptidase_M16", "Peptidase_S41"
]
no_peptidase = True
for pepdom in peptidase_domains:
if pepdom in domains:
score += 1
tabs.append(1)
no_peptidase = False
else:
tabs.append(0)
# cluster has transporter
transport_domains = ["PF00005", "PF00664"]
for transpdom in transport_domains:
if transpdom in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has response regulator (PF00072)
if "PF00072" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has major facilitator (PF07690)
if "PF07690" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has ATPase (PF13304)
if "PF13304" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has Fer4_12 (PF13353)
if "PF13353" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# cluster has rSAM (PF04055)
if "PF04055" in domains or "TIGR03975" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# cluster has no recognized peptidase
if no_peptidase:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# C-terminal portion is < 0.35 or > 0.65; C-terminal portion is defined as
# the part from the last cysteine in the last identified Cx(n)C motif to the C-terminus
# the binary opposite is also included as the next field
last_motif_c = 0
index = -1
for aa in reversed(precursor):
if aa == "C" and "C" in precursor[index - 6:index]:
last_motif_c = index + 1
index -= 1
if 0.35 <= last_motif_c / len(precursor) <= 0.65:
score += 3
tabs += [0, 1]
else:
score -= 2
tabs += [1, 0]
# SS profile count > 1
# is there more than one Cx..C structure in the sequence
cysrex = '(?=(C.{%d,%d}C))' % (CHAIN_LOWER, CHAIN_UPPER)
rex4 = re.compile(cysrex)
if len(rex4.findall(core)) > 1:
score += 2
tabs.append(1)
else:
tabs.append(0)
return score, tabs, hitends
def generate_rodeo_svm_csv(leader: str, core: str, previously_gathered_tabs: List[float]) -> List[float]:
"""Generates all the items for one candidate precursor peptide"""
precursor = leader + core
columns: List[float] = []
# Precursor Index
columns.append(1)
# classification
columns.append(0)
columns.extend(previously_gathered_tabs)
stats = ThioStatistics(core)
# Number repeating blocks of heterocyclizable residues in core
columns.append(stats.block_repeats)
# Number of core repeating Cys
columns.append(stats.c_repeats)
# Number of core repeating Ser
columns.append(stats.s_repeats)
# Number of core repeating Thr
columns.append(stats.t_repeats)
# Number of blocks of heterocyclizable residues in core
columns.append(stats.heteroblocks)
# Average core heterocycle block length
columns.append(stats.average_heteroblock_length)
# Precursor peptide mass (unmodified)
columns.append(utils.RobustProteinAnalysis(leader+core, monoisotopic=True).molecular_weight())
# Unmodified leader peptide mass
columns.append(utils.RobustProteinAnalysis(leader, monoisotopic=True).molecular_weight())
# Unmodified core peptide mass
columns.append(utils.RobustProteinAnalysis(core, monoisotopic=True).molecular_weight())
# Length of Precursor
columns.append(len(precursor))
# Length of Leader
columns.append(len(leader))
# Length of Core
columns.append(len(core))
# Ratio of length of leader / length of core
columns.append(len(core) / len(leader))
# Ratio of heterocyclizable residues / length of core
columns.append(sum([core.count(aa) for aa in "CST"]) / len(core))
# Number in leader of each amino acid
columns += [leader.count(aa) for aa in "ARDNCQEGHILKMFPSTWYV"]
# Aromatics in leader
columns.append(sum([leader.count(aa) for aa in "FWY"]))
# Neg charged in leader
columns.append(sum([leader.count(aa) for aa in "DE"]))
# Pos charged in leader
columns.append(sum([leader.count(aa) for aa in "RK"]))
# Charged in leader
columns.append(sum([leader.count(aa) for aa in "RKDE"]))
# Aliphatic in leader
columns.append(sum([leader.count(aa) for aa in "GAVLMI"]))
# Hydroxyl in leader
columns.append(sum([leader.count(aa) for aa in "ST"]))
# Counts of AAs in core
columns += [core.count(aa) for aa in "ARDNCQEGHILKMFPSTWYV"]
# Aromatics in core
columns.append(sum([core.count(aa) for aa in "FWY"]))
# Neg charged in core
columns.append(sum([core.count(aa) for aa in "DE"]))
# Pos charged in core
columns.append(sum([core.count(aa) for aa in "RK"]))
# Charged in core
columns.append(sum([core.count(aa) for aa in "RKDE"]))
# Aliphatic in core
columns.append(sum([core.count(aa) for aa in "GAVLMI"]))
# Hydroxyl in core
columns.append(sum([core.count(aa) for aa in "ST"]))
# Counts of AAs in entire precursor (leader+core)
columns += [precursor.count(aa) for aa in "ARDNCQEGHILKMFPSTWYV"]
# Aromatics in precursor
columns.append(sum([precursor.count(aa) for aa in "FWY"]))
# Neg charged in precursor
columns.append(sum([precursor.count(aa) for aa in "DE"]))
# Pos charged in precursor
columns.append(sum([precursor.count(aa) for aa in "RK"]))
# Charged in precursor
columns.append(sum([precursor.count(aa) for aa in "RKDE"]))
# Aliphatic in precursor
columns.append(sum([precursor.count(aa) for aa in "GAVLMI"]))
# Hydroxyl in precursor
columns.append(sum([precursor.count(aa) for aa in "ST"]))
return columns
def acquire_rodeo_heuristics(leader: str, core: str, # pylint: disable=too-many-branches,too-many-statements
domains: Set[str]) -> Tuple[int, List[float]]:
""" Calculate heuristic scores for RODEO """
tabs: List[float] = []
score = 0
# Contains TOMM YcaO (PF02624)
if "YcaO" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Contains LanB N-terminal domain (PF04738)
if "Lant_dehyd_N" in domains or "PF04738" in domains or "tsrC" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Contains LanB C-terminal domain (PF14028)
if "Lant_dehyd_C" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Contains TOMM dehydrogenase (PF00881)
if "PF00881" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Contains rSAM methyltransferase (PF04055)
if "PF04055" in domains:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Contains P450 (PF00067)
if "p450" in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Contains ABC transporter or abhydrolase
abc_transp_abhydrolases = ["PF00005", "PF01061", "PF12698", "PF12697", "PF00561"]
for dom in abc_transp_abhydrolases:
if dom in domains:
score += 1
tabs.append(1)
else:
tabs.append(0)
# CSS/CTT, SS/SSS/SSS, CC/CCC/CCCC, TT/TT/TTTT motifs
motifs = (('CS{2,}', 1), ('CT{2,}', 1),
('S{2,}', 1), ('S{3,}', 1), ('S{4,}', 2),
('C{2,}', 1), ('C{3,}', 1), ('C{4,}', 2),
('T{2,}', 1), ('T{3,}', 1), ('T{4,}', 2))
for motif in motifs:
if re.search(motif[0], core):
score += motif[1]
tabs.append(1)
else:
tabs.append(0)
# No Cys/Ser/Thr core residues
for aa in "CST":
if aa not in core:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Mass of core peptide (unmodified) < 2100
core_analysis = utils.RobustProteinAnalysis(core, monoisotopic=True)
if core_analysis.molecular_weight() < 2100:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Sum of repeating Cys/Ser/Thr > 4
stats = ThioStatistics(core)
if stats.cst_repeats > 4:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Avg heterocycle block length > 3
if stats.average_heteroblock_length > 3:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Leader net charge < 5
charge_dict = {"E": -1, "D": -1, "K": 1, "R": 1}
leader_charge = sum([charge_dict[aa] for aa in leader if aa in charge_dict])
if leader_charge < 5:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Leader net charge > 0
if leader_charge > 0:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Leader contains a Cys
if "C" in leader:
score -= 1
tabs.append(1)
else:
tabs.append(0)
# Peptide terminates Cys/Ser/Thr
if core[-1] in "CST":
score += 1
tabs.append(1)
else:
tabs.append(0)
# Core contains >= 2 positive residues
if sum([core.count(aa) for aa in "RK"]) >= 2:
score -= 1
tabs.append(1)
else:
tabs.append(0)
# Number of heterocyclizable residues to core ratio > 0.4
if sum([core.count(aa) for aa in "CST"]) / len(core) >= 0.4:
score += 2
tabs.append(1)
else:
tabs.append(0)
return score, tabs
def acquire_rodeo_heuristics(record: Record, cluster: Cluster,
query: CDSFeature, leader: str,
core: str) -> Tuple[int, List[Union[float, int]]]:
"""Calculate heuristic scores for RODEO"""
tabs = [] # type: List[Union[float, int]]
score = 0
# Calcd. lasso peptide mass (Da) (with Xs average out)
core_analysis = utils.RobustProteinAnalysis(core,
monoisotopic=True,
ignore_invalid=False)
tabs.append(float(core_analysis.molecular_weight()))
# Distance to any biosynthetic protein (E, B, C)
hmmer_profiles = ['PF13471', 'PF00733', 'PF05402']
distance = utils.distance_to_pfam(record, query, hmmer_profiles)
tabs.append(distance)
# Within 500 nucleotides of any biosynthetic protein (E, B, C) +1
if distance < 500:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Within 150 nucleotides of any biosynthetic protein (E, B, C) +1
if distance < 150:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Greater than 1000 nucleotides from every biosynthetic protein (E, B, C) -2
if distance > 1000:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Core region has 2 or 4 Cys residues +1
if core.count("C") in [2, 4]:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader region is longer than core region +2
if len(leader) > len(core):
score += 2
tabs.append(1)
else:
tabs.append(0)
# Core has 7 (Glu) or 8(Glu/Asp) or 9 (Asp) membered ring possible +1
if 'E' in core[6:8] or 'D' in core[7:9]:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader region contains GxxxxxT +3
if re.search('(G[ARNDBCEQZGHILKMFPSTWYV]{5}T)', leader):
score += 3
tabs.append(1)
else:
tabs.append(0)
# Core starts with G +2
if core.startswith("G"):
score += 2
tabs.append(1)
else:
tabs.append(0)
# Peptide and lasso cyclase are on same strand +1
if is_on_same_strand_as(cluster, query, 'PF00733'):
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader/core region length ratio < 2 and > 0.5 +1
if 0.5 <= len(leader) / len(core) <= 2:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Core starts with Cys and has an even number of Cys 0
if core.startswith("C") and core.count("C") % 2 == 0:
score += 0
tabs.append(1)
else:
tabs.append(0)
# Core contains no Gly -4
if "G" not in core:
score -= 4
tabs.append(1)
else:
tabs.append(0)
# Core has at least one aromatic residue +1
if set("FWY") & set(core):
score += 1
tabs.append(1)
else:
tabs.append(0)
# Core has at least 2 aromatic residues +2
if sum([core.count(aa) for aa in list("FWY")]) >= 2:
score += 2
tabs.append(1)
else:
tabs.append(0)
# Core has odd number of Cys -2
if core.count("C") % 2 != 0:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Leader region contains Trp -1
if "W" in leader:
score -= 1
tabs.append(1)
else:
tabs.append(0)
# Leader region contains Lys +1
if "K" in leader:
score += 1
tabs.append(1)
else:
tabs.append(0)
# Leader region has Cys -2
if "C" in leader:
score -= 2
tabs.append(1)
else:
tabs.append(0)
# Gene cluster does not contain PF13471 -2
if utils.distance_to_pfam(record, query, ['PF13471']) == -1 or \
utils.distance_to_pfam(record, query, ['PF13471']) > 10000:
score -= 2
# Peptide utilizes alternate start codon -1
if not str(query.extract(record.seq)).startswith("ATG"):
score -= 1
return score, tabs
def molecular_weight(self) -> float:
""" Determines the weight of the core peptide """
if not self.core:
raise ValueError("Cannot calculate weights without a core")
return self._calculate_weight(
utils.RobustProteinAnalysis(self.core, monoisotopic=False))