def forms_dimer(primer1, primer2, struct_configs, heterodimer_threshold):
heterodimer = primer3.calcHeterodimer(primer1.seq, primer2.seq,
**struct_configs)
if heterodimer.structure_found and heterodimer.dg < heterodimer_threshold:
return True
heterodimer2 = primer3.calcHeterodimer(primer1.seq, primer2.revcomp,
**struct_configs)
if heterodimer2.structure_found and heterodimer2.dg < heterodimer_threshold:
return True
return False
def SelfDimer(Seq):
TemResult = primer3.calcHeterodimer(Seq, Seq)
## print (TemResult.dg/1000)
if abs(TemResult.dg / 1000) > SelfDimerDeltaG:
return 0
else:
return 1
def findBestCommonPrimerIn3pEndRange(common_primer_end3p_range, strand,
discriminatory_primer, genome_str,
ref_genome_str, idx_lut, edge_lut,
mismatch_lut, primer_finder_params):
"""Finds best candidate primer 3p given range of ends to search over.
Returns:
primercandidate.CandidatePrimer, or None.
"""
best_common_primer_candidate = None
for j in common_primer_end3p_range:
candidate = primercandidate.findCommonPrimer(j, strand, idx_lut,
genome_str,
ref_genome_str, edge_lut,
mismatch_lut,
primer_finder_params)
if candidate is not None:
heterodimer_tm = primer3.calcHeterodimer(
discriminatory_primer.seq, candidate.seq,
**primer_finder_params['thermo_params']).tm
if heterodimer_tm > primer_finder_params['spurious_tm_clip']:
continue
if best_common_primer_candidate is None:
best_common_primer_candidate = candidate
elif best_common_primer_candidate.score < candidate.score:
best_common_primer_candidate = candidate
return best_common_primer_candidate
def evaluate(self, problem):
if not PRIMER3_AVAILABLE:
raise ImportError(
"Using avoid_heterodimerization requires primer3"
" installed (pip install primer3-py)"
)
if len(self.other_primers_sequences) == 0:
return SpecEvaluation(
specification=self,
problem=problem,
score=0,
locations=[self.location],
message="No existing primer"
)
sequence = self.location.extract_sequence(problem.sequence)
melting_temps = [
primer3.calcHeterodimer(sequence, other_seq).tm
for other_seq in self.other_primers_sequences
]
largest_tm = max(melting_temps)
# hackish penalty to guide optimization:
penalty = 0.001 * sum(melting_temps) / len(melting_temps)
score = self.tmax - largest_tm - penalty
return SpecEvaluation(
specification=self,
problem=problem,
score=score,
locations=[self.location],
message="Largest Tm = %.1f " % largest_tm,
)
def findBestCommonPrimerIn3pEndRange(
common_primer_end3p_range, strand, discriminatory_primer,
genome_str, ref_genome_str, idx_lut, edge_lut, mismatch_lut,
primer_finder_params):
"""Finds best candidate primer 3p given range of ends to search over.
Returns:
primercandidate.CandidatePrimer, or None.
"""
best_common_primer_candidate = None
for j in common_primer_end3p_range:
candidate = primercandidate.findCommonPrimer(
j, strand, idx_lut, genome_str, ref_genome_str, edge_lut,
mismatch_lut, primer_finder_params)
if candidate is not None:
heterodimer_tm = primer3.calcHeterodimer(
discriminatory_primer.seq,
candidate.seq,
**primer_finder_params['thermo_params']).tm
if heterodimer_tm > primer_finder_params['spurious_tm_clip']:
continue
if best_common_primer_candidate is None:
best_common_primer_candidate = candidate
elif best_common_primer_candidate.score < candidate.score:
best_common_primer_candidate = candidate
return best_common_primer_candidate
def primer3_check_heterodimer(primer1, primer2):
thermoResult = primer3.calcHeterodimer(primer1, primer2)
dg = thermoResult.dg / 1000
if (dg < -10):
return False
else:
return True
def primer3_check_reverse_compliments(primer1, primer2):
rprimer1 = reverse_complement(primer1)
thermoResult = primer3.calcHeterodimer(primer2, rprimer1)
dg = thermoResult.dg / 1000
if (dg < -10):
return False
else:
return True
def pcalcHeteDimer(seq1, seq2):
res = primer3.calcHeterodimer(seq1, seq2)
dg = "{0:.2f}".format(res.dg)
tm = "{0:.2f}".format(res.tm)
if res.structure_found == True and float(dg) < -1:
return "{0}{1}".format("HeteDimer:",
"+:tm:" + tm + ";deltaG:" + dg)
else:
return ""
def find_structures(cls,
folder,
seq1,
seq2=None,
sodium=0.05,
magnesium=0.0,
temperature=25,
concentration=0.00000025,
**kwargs):
"""
Should return the list of 'Structure' objects with delta-G, deltaH, deltaS, and Tm values.
Accepts 1 or 2 input sequences. Automatically runs either:
* Hairpin (1 input sequence: A=seq1, UNAFold run on A)
* Homodimer (2 identical input sequences: A=seq1=seq2, UNAFold run on A & A)
* Heterodimer (2 input sequences: A=seq1 B=seq2, UNAFold run on A & B)
"""
import primer3
mv_conc = sodium * 1000 # 50.0 # in mM
dv_conc = magnesium * 1000 # 0.0 # in mM
dntp_conc = 0.6 # in mM
dna_conc = concentration * 1000 * 1000 * 1000 # 250.0 # in nM
#temperature = 25 # keep as-is
if (seq1 == seq2): # Homodimer calculation
t = primer3.calcHomodimer(seq1,
mv_conc=mv_conc,
dv_conc=dv_conc,
dntp_conc=dntp_conc,
dna_conc=dna_conc,
temp_c=temperature)
elif (seq2 == None): # Hairpin calculation
t = primer3.calcHairpin(seq1,
mv_conc=mv_conc,
dv_conc=dv_conc,
dntp_conc=dntp_conc,
dna_conc=dna_conc,
temp_c=temperature)
else: # Heterodimer calculation, Tm calculation [seq1, rc(seq1)]
t = primer3.calcHeterodimer(seq1,
seq2,
mv_conc=mv_conc,
dv_conc=dv_conc,
dntp_conc=dntp_conc,
dna_conc=dna_conc,
temp_c=temperature)
if t.structure_found:
s = Structure(seq1, seq2, t.dg / 1000, t.dh / 1000, t.ds, t.tm,
sodium, magnesium, temperature, concentration)
else:
s = Structure(seq1, seq2, math.inf, math.inf, math.inf, math.inf,
sodium, magnesium, temperature, concentration)
return [s]
def thermo(self):
return {
"heterodimer":
primer3.calcHeterodimer(self.p1._safe_sequence,
self.p2._safe_sequence),
"left":
self.p1.thermo(),
"right":
self.p2.thermo(),
}
def heterodimer_dg(seq1, seq2, mv_cation=0, primer_conc=0):
dg = (primer3.calcHeterodimer(seq1,
seq2,
mv_conc=mv_cation,
dv_conc=0,
dntp_conc=0,
dna_conc=primer_conc,
temp_c=60,
max_loop=30)).tm
return float(("{0:.2f}".format(round(dg, 2))))
def DeltaGsum(list):
Gsum = 0
for i in range(0, len(list)):
for j in range(0, len(list)):
if (i != j):
if (i == 1 or i == 4):
Tem1 = primer3.calcHeterodimer(list[i][:ParametersR],
list[j])
DeltaGTem1 = Tem1.dg
Tem2 = primer3.calcHeterodimer(list[i][-ParametersR:],
list[j])
DeltaGTem2 = Tem2.dg
Gsum = DeltaGTem1 + DeltaGTem2 + Gsum
else:
Tem = primer3.calcHeterodimer(list[i][-ParametersR:],
list[j])
DeltaGTem = Tem.dg
Gsum = DeltaGTem + Gsum
return (Gsum / 1000)
def max_dg(self, **kwargs) -> float:
if self.length > 60:
return 0
configs = Primer.STRUCT_CONFIGS.copy()
if self.conc is not None:
configs["dna_conc"] = self.conc
return round(
primer3.calcHeterodimer(self.seq, self.revcomp, **configs).dg, 1)
def _cross_dimer_check(self, seq_tuple):
package, oligo_a, oligo_b, seq1, seq2 = seq_tuple
try:
tr = primer3.calcHeterodimer(seq1, seq2, mv_conc = self.primer_monovalent_cations, dv_conc = self.primer_divalent_cations, dntp_conc = self.primer_dntps, dna_conc = self.primer_annealing_oligo, temp_c = self.primer_annealing_temp)
deltaG = tr.dg
Tm = tr.tm
except:
deltaG = "NaN"
Tm = "NaN"
df = pd.DataFrame(columns=['Package', 'Primer1', 'Primer2', 'dG', 'Tm'])
df.loc[len(df)] = [str(package), str(oligo_a), str(oligo_b), str(deltaG), str(Tm)]
return df
def DeltaGmax(list):
Gmax = []
for i in range(0, len(list)):
for j in range(0, len(list)):
if (i != j):
if (i == 1 or i == 4):
Tem1 = primer3.calcHeterodimer(list[i][:ParametersR],
list[j])
DeltaGTem1 = Tem1.dg
Tem2 = primer3.calcHeterodimer(list[i][-ParametersR:],
list[j])
DeltaGTem2 = Tem2.dg
Gmax.append(abs(DeltaGTem1))
Gmax.append(abs(DeltaGTem2))
else:
Tem = primer3.calcHeterodimer(list[i][-ParametersR:],
list[j])
DeltaGTem = Tem.dg
Gmax.append(abs(DeltaGTem))
return max(Gmax) / 1000
def get_misprime_tm_inner(self, seq1, seq2, temp):
heterodimer = primer3.calcHeterodimer(seq1,
seq2,
mv_conc=self.mv_conc,
dv_conc=self.dv_conc,
dntp_conc=self.dntp_conc,
dna_conc=self.dna_conc,
temp_c=temp,
max_loop=MAX_LOOP)
if not heterodimer.structure_found:
return -100.
return heterodimer.tm
def primer_pairing_inner(primer_position_index_result,
primer_reverse_complement_position_index_result,
min_amplicon_length,
max_amplicon_length,
reference_sequence,
Tm=47):
primer_ID = []
primer = []
primer_reverse_complement_ID = []
primer_reverse_complement = []
amplicons_length = []
amplicon = []
for ID, primer_seq, position in zip(
primer_position_index_result[0],
primer_position_index_result[1],
primer_position_index_result[2]): # 正向引物信息提取
for reverse_complement_ID, reverse_complement_primer_seq, reverse_complement_position in zip(
primer_reverse_complement_position_index_result[0],
primer_reverse_complement_position_index_result[1],
primer_reverse_complement_position_index_result[2]
): # 反向引物信息提取
amplicon_length = int(reverse_complement_position) - int(
position) # 产物长度计算
if int(min_amplicon_length) <= amplicon_length <= int(
max_amplicon_length):
# 在产物长度满足限制条件时，检查正反向引物相互作用
heterodimer = primer3.calcHeterodimer(
primer_seq,
reverse_complement_primer_seq,
mv_conc=50.0,
dv_conc=3,
dna_conc=200,
temp_c=25)
heterodimer_Tm = str(heterodimer).split('tm=')[1].split(
',')[0]
if float(heterodimer_Tm) < float(Tm):
primer_ID.append(str(ID))
primer.append(str(primer_seq))
primer_reverse_complement_ID.append(
str(reverse_complement_ID))
primer_reverse_complement.append(
str(reverse_complement_primer_seq))
amplicons_length.append(str(amplicon_length))
amplicon.append(reference_sequence[
position:reverse_complement_position])
primer_pairing_inner_result = [
primer_ID, primer, primer_reverse_complement_ID,
primer_reverse_complement, amplicon, amplicons_length
]
return primer_pairing_inner_result
def checkSetForHeterodimers(set_of_primer_sets, tm_max=40):
all_primers = []
for bin_idx, primer_pair_idx in enumerate(set_of_primer_sets):
primer_set_obj = combined_bins[bin_idx][primer_pair_idx]
all_primers.append((bin_idx, primer_set_obj.d_primer))
all_primers.append((bin_idx, primer_set_obj.w_primer))
all_primers.append((bin_idx, primer_set_obj.c_primer))
for p1, p2 in itertools.combinations(all_primers, 2):
if (primer3.calcHeterodimer(p1[1].seq, p2[1].seq, **
thermo_params).tm > tm_max):
return (False, p1[0], p2[0])
return (True, )
def analyze_heterostructures(primers,
structures,
heterodimer_threshold=-5000,
mv_conc=None,
dv_conc=None):
for primer1, primer2 in itertools.combinations(primers, 2):
# At least one primer needs to be shorter than 60 bp.
if len(primer1) > 60 and len(primer2) > 60:
logging.warning(
f"Primers {primer1.name} and {primer2.name} are too long (> 60 bp) for analysis of "
f"heterostructures. ")
continue
keyword_args = {
"mv_conc": mv_conc,
"dv_conc": dv_conc,
"dna_conc": max(primer1.conc, primer2.conc),
}
heterodimer = primer3.calcHeterodimer(primer1.seq, primer2.seq,
**keyword_args)
if heterodimer.structure_found and heterodimer.dg < heterodimer_threshold:
structures.add_struture(
f"{primer1.name} and {primer2.name}",
"Heterodimer",
round(heterodimer.dg, 1),
round(heterodimer.tm, 1),
)
heterodimer2 = primer3.calcHeterodimer(primer1.seq, primer2.revcomp,
**keyword_args)
if heterodimer2.structure_found and heterodimer2.dg < heterodimer_threshold:
structures.add_struture(
f"{primer1.name} and {primer2.name}'",
"Heterodimer",
round(heterodimer2.dg, 1),
round(heterodimer2.tm, 1),
)
def process(configd, desc_d, name, d_pos):
l_f = sorted(set(d_pos['f']))
l_r = sorted(set(d_pos['r']))
a_f = numpy.array(l_f)
a_r = numpy.array(l_r)
range_st = range(configd['LEN_LOWER'], configd['LEN_UPPER'])
for i, n in enumerate(a_f):
f_name = '{}__{}'.format(name, n)
f_d = desc_d['f'][f_name]
f = f_d['seq']
for j, m in enumerate(l_r):
diff = m - n
if diff < 0:
continue
elif diff in range_st:
r_name = '{}__{}'.format(name, m)
r_d = desc_d['r'][r_name]
r = r_d['seq']
# desc e.g:
#contiglen__184765;di__'f';tm__56.9135107847;gc__0.6
if configd['SS']:
# check heterodimer
hetero = primer3.calcHeterodimer(
f.encode('utf-8'),
r.encode('utf-8'),
)
if hetero.dg < configd['DI_DG_LIMIT']:
continue
# forward, f_tm, f_gc, reverse, r_tm, r_gc
mes = ('{}\t{}\t'
'{}\t{}\t{}\t'
'{}\t{:.1f}\t{:.2f}\t'
'{}\t{:.1f}\t{:.2f}\t')
print(
mes.format(
name,
f_d['contiglen'],
n,
m,
diff,
f,
float(f_d['tm']),
float(f_d['gc']),
r,
float(r_d['tm']),
float(r_d['gc']),
))
# break when out of range since already sorted
elif diff > configd['LEN_UPPER']:
break
def thermo(self):
settings = self._thermo_settings
if self._thermo is None:
if len(self.sequence) > 60:
warning = "sequence length greater than 60. Thermo results are limited to 60bp."
else:
warning = ""
self._thermo = {
"hairpin":
primer3.calcHairpin(self._safe_sequence, **settings),
"homodimer":
primer3.calcHomodimer(self._safe_sequence, **settings),
"annealing":
primer3.calcHeterodimer(self.anneal, rc(self.anneal),
**settings),
"sequence":
primer3.calcHeterodimer(self._safe_sequence,
rc(self._safe_sequence), **settings),
"warning":
warning,
}
return self._thermo
def _calc_thermal_parameters(self, seq1, seq2, seqtype):
if seqtype == "Primer" or seqtype == "Product":
mv, dv, dntp, dna, temp = self.primer_monovalent_cations, self.primer_divalent_cations, self.primer_dntps, self.primer_annealing_oligo, self.primer_annealing_temp
else:
mv, dv, dntp, dna, temp = self.probe_monovalent_cations, self.probe_divalent_cations, self.probe_dntps, self.probe_annealing_oligo, self.probe_annealing_temp
try:
tr = primer3.calcHeterodimer(str(seq1), str(seq2), mv_conc = mv, dv_conc = dv, dntp_conc = dntp, dna_conc = dna, temp_c = temp)
deltaG = round(tr.dg / 1000, 2)
Tm = round(tr.tm, 2)
except:
deltaG = "NaN"
Tm = "NaN"
return Tm, deltaG
def checkDimerEnd(primer1,primer2=None):
if not primer2:
primer2=primer1
primersDimer=round(primer3.calcHeterodimer(primer1,
primer2,
mv_conc=args.mvConc,
dv_conc=args.dvConc).dg/1000,2)
primer1End_rc=revComplement(primer1[-5:])
primer2End_rc=revComplement(primer2[-5:])
if (primersDimer<-3 and
(primer1End_rc in primer2[1:] or
primer2End_rc in primer1[1:])):
return(str(primersDimer)+'*')
return(primersDimer)
def calcDimer(seq1,
seq2,
dg_Threshold=8,
mv_conc=50.0,
dv_conc=1.5,
dntp_conc=0.25,
dna_conc=50.0,
temp_c=37,
max_loop=30):
thermoresult = primer3.calcHeterodimer(seq1, seq2, mv_conc, dv_conc,
dntp_conc, dna_conc, temp_c,
max_loop)
if math.fabs(thermoresult.dg / 1000) >= dg_Threshold:
return thermoresult.tm, thermoresult.dg / 1000
else:
return -1, -1
def calc_dimer(df):
'''
Calculate the dimer dG between each probe and each other probe. Add the
min dG to the dataframe, along with the index of the dimer partner.
'''
df['index_num'] = df.index
a = df[['index_num', 'sequence']].to_numpy()
max_ints = []
for i in range(0, len(a)):
l = []
for j in range(0, len(a)):
#This includes both homodimers and heterodimers.
l.append((primer3.calcHeterodimer(a[i][1], a[j][1], mv_conc = 300).dg/1000, a[j][0]))
maxinteraction = sorted(l, key = lambda x: x[0])[0]
max_ints.append(maxinteraction)
dimer_dG = pd.DataFrame(max_ints, index = df.index)
return dimer_dG
def judge_two_oligo(oligo_pair):
oligo_1 = oligo_pair['oligo_1']
oligo_2 = oligo_pair['oligo_2']
min_Tm = min(oligo_1['Tm'], oligo_2['Tm'], oligo_pair['min_Tm'])
Heterodimer = primer3.calcHeterodimer(oligo_1['seq'],
oligo_2['seq'],
output_structure=True)
if Heterodimer.tm > min_Tm:
return ([
oligo_1, oligo_2,
round(Heterodimer.tm, 2), Heterodimer.ascii_structure
])
EndStability = primer3.bindings.calcEndStability(oligo_1['seq'],
oligo_2['seq'])
if EndStability.tm > min_Tm:
return ([oligo_1, oligo_2, round(EndStability.tm, 2), ''])
return None
def heterodimer(self, primer: str, other_primer: str) -> float:
"""
Cached heterodimer computation with Primer3 library.
:param primer: [str] primer sequence
:return: melting temperature
"""
if len(primer) > 0:
key = primer + '_' + other_primer
if self.cached:
if key in self.cache_hetero:
return self.cache_hetero[key]
temp = calcHeterodimer(primer, other_primer, self.mv, self.dv,
self.dntp).tm
if self.cached:
self.cache_hetero[key] = temp
return temp
else:
return 0
def __init__(self, oligomer, target):
self.oligomer = oligomer
self.target = target
thermo = calcHeterodimer(self.oligomer, self.target,
output_structure=True)
self.tm = thermo.tm
self.dg = thermo.dg
self.dh = thermo.dh
self.ds = thermo.ds
# Structure and structure lines are given in the primer3 format
self.structure = thermo.ascii_structure
self.structure_lines = thermo.ascii_structure_lines
# Intended for use by getters, setters, and 'magic' properties
self._formatted_structure_lines = None
self._formatted_structure = None
self._reduced_structure = None
def judge_two_site(site_pair):
if global_var.stop_run is True:
return []
dimers = []
for (rank_1, primers_1) in site_pair['site_1'].items():
for (rank_2, primers_2) in site_pair['site_2'].items():
for (seq_1, seq_2) in product([primers_1['seq_L'], primers_1['seq_R']], [primers_2['seq_L'], primers_2['seq_R']]):
min_Tm = min(primers_1['min_Tm'], primers_2['min_Tm'])-site_pair['Tm_diff']
Heterodimer = primer3.calcHeterodimer(seq_1, seq_2, output_structure=True)
if Heterodimer.tm > min_Tm:
dimers.append([site_pair['id_1'], site_pair['id_2'], rank_1, rank_2, seq_1, seq_2, Heterodimer.tm, Heterodimer.ascii_structure])
break
EndStability = primer3.bindings.calcEndStability(seq_1, seq_2)
if EndStability.tm > min_Tm:
dimers.append([site_pair['id_1'], site_pair['id_2'], rank_1, rank_2, seq_1, seq_2, EndStability.tm, ''])
break
return dimers
def test_primers_collection_example():
def create_new_primer(existing_primers):
"""Create a new primer based on the primers created so far"""
problem = DnaOptimizationProblem(
sequence=random_dna_sequence(length=20),
constraints=[
AvoidHeterodimerization(existing_primers, tmax=3),
AvoidPattern("3x3mer"),
AvoidPattern("4xG"),
],
objectives=[EnforceGCContent(target=0.6)],
logger=None,
)
problem.resolve_constraints()
problem.optimize()
return problem.sequence
# MAIN LOOP, WHERE PRIMERS ARE CREATED ONE BY ONE
existing_primers = []
for i in range(10):
new_primer = create_new_primer(existing_primers)
existing_primers.append(new_primer)
print("PRIMERS GENERATED: \n\n%s\n" % "\n".join(existing_primers))
for sequence in existing_primers:
assert "GGGG" not in sequence
assert "CCCC" not in sequence
max_tm = max(
primer3.calcHeterodimer(seq1, seq2).tm
for seq1, seq2 in itertools.combinations(existing_primers, 2))
assert max_tm < 3
gc_contents = [gc_content(p) for p in existing_primers]
assert min(gc_contents) > 0.55
assert max(gc_contents) < 0.65
#### Reverse complement the primers and test for criterias for TSO compatibility by Primer3. Select only the primers with Tm > 50. Remove any primers with "CC" or "TTT".
rc_pr_list = []
for i in full_list:
if ("CC" not in rc(i) and "TTT" not in rc(i)
and primer3.calcTm(rc(i)) > 50):
l = OH + rc(i)
rc_pr_list.append(l)
#### Select the primers with least tendency to form heterodimers with TSO. dG > -3000 was chosen acccording to Fabio's DENV2 primer.
dg_3000 = []
for i in rc_pr_list:
result = primer3.calcHeterodimer(i, TSO)
pin = primer3.calcHairpin(i)
if result.dg > -3000:
#print(i, result.tm,result.dg, primer3.calcTm(i[23:]))
#Check for the formation of hairpins.
#print(i, primer3.calcHairpin(i))
tttt = [
i, primer3.calcTm(i[23:]), result.tm, result.dg, pin.tm, pin.dg
]
dg_3000.append(tttt)
dg_3000 = pd.DataFrame(np.array(dg_3000),
columns=[
"Primer", "Annealing Tm", "HeteroDimer Tm",
"HeteroDimer dG", "Hairpin Tm", "Hairpin dG"
])
dg_3000.iloc[:, 1:] = dg_3000.iloc[:, 1:].astype(float).round(2)
def heterodimer_dg(seq1, seq2, mv_cation=0,primer_conc=0):
dg = (primer3.calcHeterodimer(seq1, seq2,mv_conc=mv_cation, dv_conc=0, dntp_conc=0, dna_conc=primer_conc, temp_c=60, max_loop=30)).tm
return float(("{0:.2f}".format(round(dg,2))))
