class TestPeptide(TestCase):
def setUp(self):
self.simple = Peptide("SYFPEITHI")
self.gcg_ps = "MKSIYFVAGLFVMLVQGSWQRSLQDTEEKSRSFSASQADPLSDPDQMNEDKRHSQGTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAKRHDEFERHAEGTFTSDVSSYLEGQAAKEFIAWLVKGRGRRDFPEEVAIVEELGRRHADGSFSDEMNTILDNLAARDFINWLIQTKITDRK"
self.gcg_t1 = Transcript("", transcript_id="GLUC_HUMAN")
gcg_p1 = Protein(self.gcg_ps,
transcript_id='GLUC_HUMAN',
orig_transcript=self.gcg_t1)
self.w_p = Peptide("PROTEIN", {gcg_p1: [0]})
self.gcg_p1 = gcg_p1
self.gcg_v1 = Variant(
"rs5650", VariationType.SNP, 2, 162145588, 'G', 'T', {
"GLUC_HUMAN":
MutationSyntax("GLUC_HUMAN", 344, 115, "c.344C>A", "p.A115D")
}, False, False)
gcg_p1_copy = copy.deepcopy(gcg_p1)
gcg_p1_copy.vars = {0: [self.gcg_v1]}
self.w_v = Peptide("VARIANT", {gcg_p1_copy: [0]})
#self.gcg_ts = "gcatagaatgcagatgagcaaagtgagtgggagagggaagtcatttgtaacaaaaactcattatttacagatgagaaatttatattgtcagcgtaatatctgtgaggctaaacagagctggagagtatataaaagcagtgcgccttggtgcagaagtacagagcttaggacacagagcacatcaaaagttcccaaagagggcttgctctctcttcacctgctctgttctacagcacactaccagaagacagcagaaatgaaaagcatttactttgtggctggattatttgtaatgctggtacaaggcagctggcaacgttcccttcaagacacagaggagaaatccagatcattctcagcttcccaggcagacccactcagtgatcctgatcagatgaacgaggacaagcgccattcacagggcacattcaccagtgactacagcaagtatctggactccaggcgtgcccaagattttgtgcagtggttgatgaataccaagaggaacaggaataacattgccaaacgtcacgatgaatttgagagacatgctgaagggacctttaccagtgatgtaagttcttatttggaaggccaagctgccaaggaattcattgcttggctggtgaaaggccgaggaaggcgagatttcccagaagaggtcgccattgttgaagaacttggccgcagacatgctgatggttctttctctgatgagatgaacaccattcttgataatcttgccgccagggactttataaactggttgattcagaccaaaatcactgacaggaaataactatatcactattcaagatcatcttcacaacatcacctgctagccacgtgggatgtttgaaatgttaagtcctgtaaatttaagaggtgtattctgaggccacattgctttgcatgccaataaataaattttcttttagtgttgtgtagccaaaaattacaaatggaataaagttttatcaaaatattgctaaaatatcagctttaaaatatgaaagtgctagattctgttattttcttcttattttggatgaagtaccccaacctgtttacatttagcgataaaattatttttctatgatataatttgtaaatgtaaattattccgatctgacatatctgcattataataataggagaatagaagaactggtagccacagtggtgaaattggaaagagaactttcttcctgaaacctttgtcttaaaaatactcagctttcaatgtatcaaagatacaattaaataaaattttcaagcttctttaccattgtct"
##gcg_t1 = Transcript(gcg_ts, "NM_002054.4", {344: gcg_v1})
#gcg_t1 = Transcript(self.gcg_ts, 'GLUC_HUMAN', "NM_002054.4", [self.gcg_v1])
#self.w_t = Peptide("TRANSCRIPT", {gcg_p1: [0]})
def test_consistency(self):
"""
tests all __*__ (including init)
test has several asserts! If one fails, the following will not be evaluated!
"""
self.assertTrue(repr(self.simple) == "PEPTIDE:\n SYFPEITHI")
self.assertTrue(
repr(self.w_p) ==
"PEPTIDE:\n PROTEIN\nin TRANSCRIPT: GLUC_HUMAN\n\tVARIANTS:\nin PROTEIN: GLUC_HUMAN"
)
self.assertTrue(
repr(self.w_v) ==
"PEPTIDE:\n VARIANT\nin TRANSCRIPT: GLUC_HUMAN\n\tVARIANTS:\n\tVariant(g.162145588G>T)\nin PROTEIN: GLUC_HUMAN"
)
def test_getitem(self):
self.assertTrue(self.simple[1:3] == 'YF')
#TODO: document to have variant peptides from Protein with Variants use Generator
def test_get_all_variants(self):
self.assertTrue(
repr(self.w_v.get_variants_by_protein("GLUC_HUMAN")) == repr(
[self.gcg_v1]))
def test_get_all_proteins(self):
self.assertTrue(repr(self.simple.get_all_proteins()) == repr([]))
self.assertTrue(
repr(self.w_p.get_all_proteins()) == repr([self.gcg_p1]))
def test_get_all_transcripts(self):
self.assertTrue(
repr(self.w_v.get_all_transcripts()) == repr(
[Transcript(seq="", transcript_id="GLUC_HUMAN")]))
self.assertTrue(
repr(self.w_p.get_all_transcripts()) == repr([self.gcg_t1]))
def setUp(self):
self.peptides_mhcI = [Peptide("SYFPEITHI"), Peptide("IHTIEPFYS")]
self.peptides_mhcII = [Peptide("AAAAAASYFPEITHI"), Peptide("IHTIEPFYSAAAAAA")]
self.mhcI = [Allele("HLA-B*07:02"), Allele("HLA-A*02:01")]
self.mhcII = [Allele("HLA-DRB1*07:01"), Allele("HLA-DRB1*15:01")]
self.mhcII_combined_alleles = [CombinedAllele("DPA1*01:03-DPB1*01:01"), CombinedAllele("DQA1*06:02-DQB1*06:31")]
self.transcript = Transcript("")
def setUp(self):
#Peptides of different length 9,10,11,12,13,14,15
self.peptides_mhcI = [Peptide("SYFPEITHI"), Peptide("IHTIEPFYS")]
self.peptides_mhcII = [
Peptide("SYFPEITHI"),
Peptide("IHTIEPFYSAAAAAA")
]
self.mhcI = [Allele("HLA-B*15:01"), Allele("HLA-A*02:01")]
self.mhcII = [Allele("HLA-DRB1*07:01"), Allele("HLA-DRB1*15:01")]
def test_simple_assembly(self):
"""
Simple test if everything works. Solution manually tested for optimality.
:return:
"""
pred = CleavageSitePredictorFactory("PCM")
assembler = EpitopeAssembly(self.peptides, pred, solver="cbc", verbosity=0)
r = assembler.solve()
self.assertEqual(r, [Peptide("YLYDHLAPM"), Peptide("ALYDVVSTL"), Peptide("KLLPRLPGV")])
def setUp(self):
self.simple = Peptide("SYFPEITHI")
self.gcg_ps = "MKSIYFVAGLFVMLVQGSWQRSLQDTEEKSRSFSASQADPLSDPDQMNEDKRHSQGTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAKRHDEFERHAEGTFTSDVSSYLEGQAAKEFIAWLVKGRGRRDFPEEVAIVEELGRRHADGSFSDEMNTILDNLAARDFINWLIQTKITDRK"
self.gcg_t1 = Transcript("", _transcript_id="GLUC_HUMAN")
gcg_p1 = Protein(self.gcg_ps, _transcript_id='GLUC_HUMAN', _orig_transcript=self.gcg_t1)
self.w_p = Peptide("PROTEIN", {gcg_p1:[0]})
self.gcg_p1 = gcg_p1
self.gcg_v1 = Variant("rs5650", VariationType.SNP, 2, 162145588, 'G', 'T',
{"GLUC_HUMAN": MutationSyntax("GLUC_HUMAN", 344, 115, "c.344C>A", "p.A115D")}, False, False)
gcg_p1_copy = copy.deepcopy(gcg_p1)
gcg_p1_copy.vars = {0:[self.gcg_v1]}
self.w_v = Peptide("VARIANT", {gcg_p1_copy:[0]})
def compute_affinities(input_alleles, input_peptides, output_affinities,
processes, predictor):
''' Computes the binding affinities between the given peptides and HLA alleles
'''
alleles = [
Allele(a.replace('HLA-', ''))
for a in utilities.get_alleles_and_thresholds(input_alleles).index
]
LOGGER.info('Loaded %d alleles', len(alleles))
with open(input_peptides) as f:
reader = csv.DictReader(f)
peptides = [(Peptide(r['peptide']), len(r['proteins'].split(';')))
for r in reader]
peptides.sort(key=lambda p: p[1], reverse=True)
LOGGER.info('Loaded %d peptides', len(peptides))
results = utilities.parallel_apply(
get_binding_affinity_process,
((predictor.lower(), batch, alleles)
for batch in utilities.batches((p for p, _ in peptides), bsize=256)),
processes)
count = 0
for bindings in results:
bindings.to_csv(output_affinities,
header=(count == 0),
mode=('w' if count == 0 else 'a'))
count += len(bindings)
LOGGER.debug('Processed %d peptides (%.2f%%)...', count,
100 * count / len(peptides))
class TestPeptide(TestCase):
def setUp(self):
self.simple = Peptide("SYFPEITHI")
self.gcg_ps = "MKSIYFVAGLFVMLVQGSWQRSLQDTEEKSRSFSASQADPLSDPDQMNEDKRHSQGTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAKRHDEFERHAEGTFTSDVSSYLEGQAAKEFIAWLVKGRGRRDFPEEVAIVEELGRRHADGSFSDEMNTILDNLAARDFINWLIQTKITDRK"
self.gcg_t1 = Transcript("", _transcript_id="GLUC_HUMAN")
gcg_p1 = Protein(self.gcg_ps, _transcript_id='GLUC_HUMAN', _orig_transcript=self.gcg_t1)
self.w_p = Peptide("PROTEIN", {gcg_p1:[0]})
self.gcg_p1 = gcg_p1
self.gcg_v1 = Variant("rs5650", VariationType.SNP, 2, 162145588, 'G', 'T',
{"GLUC_HUMAN": MutationSyntax("GLUC_HUMAN", 344, 115, "c.344C>A", "p.A115D")}, False, False)
gcg_p1_copy = copy.deepcopy(gcg_p1)
gcg_p1_copy.vars = {0:[self.gcg_v1]}
self.w_v = Peptide("VARIANT", {gcg_p1_copy:[0]})
#self.gcg_ts = "gcatagaatgcagatgagcaaagtgagtgggagagggaagtcatttgtaacaaaaactcattatttacagatgagaaatttatattgtcagcgtaatatctgtgaggctaaacagagctggagagtatataaaagcagtgcgccttggtgcagaagtacagagcttaggacacagagcacatcaaaagttcccaaagagggcttgctctctcttcacctgctctgttctacagcacactaccagaagacagcagaaatgaaaagcatttactttgtggctggattatttgtaatgctggtacaaggcagctggcaacgttcccttcaagacacagaggagaaatccagatcattctcagcttcccaggcagacccactcagtgatcctgatcagatgaacgaggacaagcgccattcacagggcacattcaccagtgactacagcaagtatctggactccaggcgtgcccaagattttgtgcagtggttgatgaataccaagaggaacaggaataacattgccaaacgtcacgatgaatttgagagacatgctgaagggacctttaccagtgatgtaagttcttatttggaaggccaagctgccaaggaattcattgcttggctggtgaaaggccgaggaaggcgagatttcccagaagaggtcgccattgttgaagaacttggccgcagacatgctgatggttctttctctgatgagatgaacaccattcttgataatcttgccgccagggactttataaactggttgattcagaccaaaatcactgacaggaaataactatatcactattcaagatcatcttcacaacatcacctgctagccacgtgggatgtttgaaatgttaagtcctgtaaatttaagaggtgtattctgaggccacattgctttgcatgccaataaataaattttcttttagtgttgtgtagccaaaaattacaaatggaataaagttttatcaaaatattgctaaaatatcagctttaaaatatgaaagtgctagattctgttattttcttcttattttggatgaagtaccccaacctgtttacatttagcgataaaattatttttctatgatataatttgtaaatgtaaattattccgatctgacatatctgcattataataataggagaatagaagaactggtagccacagtggtgaaattggaaagagaactttcttcctgaaacctttgtcttaaaaatactcagctttcaatgtatcaaagatacaattaaataaaattttcaagcttctttaccattgtct"
##gcg_t1 = Transcript(gcg_ts, "NM_002054.4", {344: gcg_v1})
#gcg_t1 = Transcript(self.gcg_ts, 'GLUC_HUMAN', "NM_002054.4", [self.gcg_v1])
#self.w_t = Peptide("TRANSCRIPT", {gcg_p1: [0]})
def test_consistency(self):
"""
tests all __*__ (including init)
test has several asserts! If one fails, the following will not be evaluated!
"""
self.assertTrue(repr(self.simple) == "PEPTIDE:\n SYFPEITHI")
self.assertTrue(repr(self.w_p) == "PEPTIDE:\n PROTEIN\nin TRANSCRIPT: GLUC_HUMAN\n\tVARIANTS:\nin PROTEIN: GLUC_HUMAN")
self.assertTrue(repr(self.w_v) == "PEPTIDE:\n VARIANT\nin TRANSCRIPT: GLUC_HUMAN\n\tVARIANTS:\n\tVariant(g.162145588G>T)\nin PROTEIN: GLUC_HUMAN")
def test_getitem(self):
self.assertTrue(self.simple[1:3] == 'YF')
#TODO: document to have variant peptides from Protein with Variants use Generator
def test_get_all_variants(self):
self.assertTrue(repr(self.w_v.get_variants_by_protein("GLUC_HUMAN")) == repr([self.gcg_v1]))
def test_get_all_proteins(self):
self.assertTrue(repr(self.simple.get_all_proteins()) == repr([]))
self.assertTrue(repr(self.w_p.get_all_proteins()) == repr([self.gcg_p1]))
def test_get_all_transcripts(self):
self.assertTrue(repr(self.w_v.get_all_transcripts()) == repr([Transcript(_seq="", _transcript_id="GLUC_HUMAN")]))
self.assertTrue(repr(self.w_p.get_all_transcripts()) == repr([self.gcg_t1]))
def read_lines(file):
peptides = []
with open(file, "r") as f:
for l in f:
if not l.startswith("#") and l.strip() != "" and not l.startswith(
"Epitope") and not l.startswith("Sequence"):
#print l, l.split()
pep = l.split()[0].strip()
peptides.append(Peptide(pep))
return peptides
def optitope(input_affinities, input_peptides, input_alleles, output_vaccine,
epitopes, min_alleles, min_proteins):
with open(input_peptides) as f:
reader = csv.DictReader(f)
peptides = {
# we don't really need the actual protein sequence, just fill it with the id to make it unique
Peptide(r['peptide']):
set(Protein(gid, gene_id=gid) for gid in r['proteins'].split(';'))
for r in reader
}
LOGGER.info('Loaded %d peptides', len(peptides))
allele_data = utilities.get_alleles_and_thresholds(input_alleles).to_dict(
'index')
thresholds = {
allele.replace('HLA-', ''): data['threshold']
for allele, data in allele_data.items()
}
LOGGER.info('Loaded %d alleles', len(thresholds))
affinities = utilities.affinities_from_csv(input_affinities,
allele_data,
peptide_coverage=peptides)
LOGGER.info('Loaded %d affinities', len(affinities))
LOGGER.info("Creating vaccine...")
model = OptiTope(affinities, thresholds, k=epitopes, solver='gurobi')
if min_alleles > 0:
model.activate_allele_coverage_const(min_alleles)
LOGGER.info('Vaccine will cover at least %f alleles', min_alleles)
if min_proteins > 0:
model.activate_antigen_coverage_const(min_proteins)
LOGGER.info('Vaccine will cover at least %f proteins', min_proteins)
vaccine = model.solve()
LOGGER.info('Vaccine summary:')
with open(output_vaccine, 'w') as f:
writer = csv.writer(f)
writer.writerow(('cocktail', 'index', 'epitope'))
total_ig = 0.0
for i, epitope in enumerate(vaccine):
writer.writerow((0, i, epitope))
epitope_immunog = sum(model.instance.p[a] *
model.instance.i[epitope, a]
for a in model.instance.A)
total_ig += epitope_immunog
LOGGER.info(' %s - %.2f', epitope, epitope_immunog)
LOGGER.info('Total immunogenicity: %.2f', total_ig)
def get_cleavage_score_process(penalty, cleavage_model, window_size, epitopes):
#predictor = CleavageSitePredictorFactory(cleavage_model)
assert cleavage_model.lower() == 'pcm'
from Fred2.CleavagePrediction import PCM
predictor = PCM()
results = []
for ep_from, ep_to in epitopes:
preds = predictor.predict(Peptide(ep_from + ep_to))
score = 0.0
join_pos = len(ep_from) - 1
half_size = int((window_size - 1) / 2)
for i, (_, lik) in enumerate(preds.values):
if i - half_size <= join_pos <= i + half_size:
weight = -1 if i == join_pos else penalty
score += weight * lik
results.append((ep_from, ep_to, score))
return results
def append_score(dt2, alleles):
"""
Given a choped sequence (output from sliding_window()),
append the immunogenicity scores
"""
peptides_to_compute = [
Peptide(peptide) for peptide in set(list(dt2["MT"]) + list(dt2["WT"]))
]
res = fred2wrap.predict_peptide_effects(peptides_to_compute, alleles)
res["peptide"] = [str(peptide) for peptide in res["peptide"]]
full = pd.merge(dt2, res, how='left', left_on="WT", right_on="peptide")
full = full.rename(columns={'score': 'WT_score'})
del full["peptide"]
full = pd.merge(full,
res,
how='left',
left_on=["MT", "method", "allele"],
right_on=["peptide", "method", "allele"])
full = full.rename(columns={'score': 'MT_score'})
del full["peptide"]
return full
def generate_epitope_result(input, allele_file):
"""
generates EpitopePredictionResult from output of epitopeprediction and neoepitopeprediction
"""
#first generate alleles in allele file
alleles = {}
with open(allele_file, "r") as af:
for l in af:
allele, freq = l.split("\t")
alleles[allele] = Allele(allele, prob=float(freq))
r_raw = pandas.read_csv(input, sep="\t")
res_dic = {}
method = r_raw.loc[0, "Method"]
columns = set(["Sequence", "Method", "Antigen ID", "Variant"])
alleles_raw = [c for c in r_raw.columns if c not in columns]
for k, row in r_raw.iterrows():
seq = row["Sequence"]
protPos = collections.defaultdict(list)
try:
protPos = {Protein(p, gene_id=p, transcript_id=p): [0] for p in str(row["Antigen ID"]).split(",")}
except KeyError:
pass
pep = Peptide(seq, protein_pos=protPos)
for a in alleles_raw:
if a in alleles:
if alleles[a] not in res_dic:
res_dic[alleles[a]] = {}
res_dic[alleles[a]][pep] = float(row[a])
if not res_dic:
sys.stderr.write("HLA alleles of population and HLA used for prediction did not overlap.")
sys.exit(-1)
df_result = EpitopePredictionResult.from_dict(res_dic)
df_result.index = pandas.MultiIndex.from_tuples([tuple((i, method)) for i in df_result.index],
names=['Seq', 'Method'])
return df_result, method
def setUp(self):
self.simple = Peptide("SYFPEITHI")
self.gcg_ps = "MKSIYFVAGLFVMLVQGSWQRSLQDTEEKSRSFSASQADPLSDPDQMNEDKRHSQGTFTSDYSKYLDSRRAQDFVQWLMNTKRNRNNIAKRHDEFERHAEGTFTSDVSSYLEGQAAKEFIAWLVKGRGRRDFPEEVAIVEELGRRHADGSFSDEMNTILDNLAARDFINWLIQTKITDRK"
self.gcg_t1 = Transcript("", transcript_id="GLUC_HUMAN")
gcg_p1 = Protein(self.gcg_ps,
transcript_id='GLUC_HUMAN',
orig_transcript=self.gcg_t1)
self.w_p = Peptide("PROTEIN", {gcg_p1: [0]})
self.gcg_p1 = gcg_p1
self.gcg_v1 = Variant(
"rs5650", VariationType.SNP, 2, 162145588, 'G', 'T', {
"GLUC_HUMAN":
MutationSyntax("GLUC_HUMAN", 344, 115, "c.344C>A", "p.A115D")
}, False, False)
gcg_p1_copy = copy.deepcopy(gcg_p1)
gcg_p1_copy.vars = {0: [self.gcg_v1]}
self.w_v = Peptide("VARIANT", {gcg_p1_copy: [0]})
def popcover(input_peptides, input_affinities, input_alleles, output_vaccine,
processes, epitopes):
with open(input_peptides) as f:
reader = csv.DictReader(f)
peptides = {
Peptide(r['peptide']): set(r['proteins'].split(';'))
for r in reader
}
LOGGER.info('Loaded %d peptides', len(peptides))
allele_data = utilities.get_alleles_and_thresholds(input_alleles).to_dict(
'index')
thresholds = {
allele.replace('HLA-', ''): data['threshold']
for allele, data in allele_data.items()
}
LOGGER.info('Loaded %d alleles', len(thresholds))
affinities = utilities.affinities_from_csv(input_affinities, allele_data,
peptides)
LOGGER.info('Loaded %d affinities', len(affinities))
LOGGER.info("Creating vaccine...")
model = PopCover(affinities,
thresholds,
k=epitopes,
processes=processes if processes > 0 else
(mp.cpu_count() + processes))
vaccine = model.solve()
with open(output_vaccine, 'w') as f:
writer = csv.writer(f)
writer.writerow(('cocktail', 'index', 'epitope'))
for i, epitope in enumerate(vaccine):
writer.writerow((0, i, epitope))
LOGGER.info(' %s', epitope)
def read_epitope_input(args, alleles, exclude):
"""
reads in epitope files generated by NGSAnalyzer+ImmogenicityPredictor
Header NGSAnalyzer:
mutation - position of the mutation in the reference genome (currently hg19); format:
chromosome_position; the position is zero-based
gene - gene affected by the mutation
transcript - transcript affected by the mutation (UCSC known genes transcript ID)
transcript_expression - expression in RPKM/FPKM of the affected transcript
neopeptide - peptide resulting from the mutation in the given transcript
length_of_neopeptide - length of the neopeptide
HLA - HLA used for the binding prediction of the neopeptide
HLA_class1_binding_prediction - predicted binding affinity (currently rank score of IEDB consensus tool)
Header ImmunogenicityPredictor:
immunogenicity - predicted immunogenicity for specific HLA allele in column
distance - distance-to-self estimation specific for HLA allele in column
[uncertainty] - If immunopredictor can estimate prediction uncertainty
:param args: Input arguments
:param alleles: HLA alleles
:param exlude: excluded peptides
:return: df_epitope - EpitopePredictionResult
distance - dict(pep_string, float)
expression - dict(gene_id, float)
uncertainty - dict(pep_string, float)
pep_to_mutation - dict(pep_string, mutation_string)
"""
distance = {}
uncertainty = {}
expression = {}
seq_to_pep = {}
gene_to_prot = {}
hla_to_allele = {a.name: a for a in alleles}
df_pred = {a: {} for a in alleles}
pep_to_mutation = {}
with open(args.input, "rU") as f:
reader = csv.DictReader(f, delimiter='\t')
for row in reader:
seq = row["neopeptide"]
if (exclude is None or seq not in exclude) and len(seq) > 0:
if seq in seq_to_pep:
pep = seq_to_pep[seq]
else:
pep = Peptide(seq.upper())
seq_to_pep[seq] = pep
try:
allele = hla_to_allele[row["HLA"].replace("HLA-", "")]
except:
logging.warning(
"HLA {allele} was not contained in the provided allele file. Please check your input."
.format(allele=row["HLA"]))
continue
gene = row["gene"]
if gene in gene_to_prot:
prot = gene_to_prot[gene]
else:
prot = Protein("", gene_id=gene, transcript_id=gene)
gene_to_prot[gene] = prot
pep.proteins[prot.transcript_id] = prot
pep.proteinPos[prot.transcript_id].append(0)
pep_to_mutation.setdefault(seq, []).append(row["mutation"])
expression.setdefault(gene, []).append(
float(row["transcript_expression"]))
if args.rank:
df_pred[allele][pep] = max(
0., 1. - float(row[args.immunogenicity]) / 100.0)
else:
df_pred[allele][pep] = max(
0., 1. -
math.log(float(row[args.immunogenicity]), 50000)
) if args.immunogenicity == "HLA_class1_binding_prediction" else float(
row[args.immunogenicity])
if args.distance is not None:
distance[(seq, allele.name)] = float(row[args.distance])
if args.uncertainty is not None:
uncertainty[(seq,
allele.name)] = float(row[args.uncertainty])
if args.taa is not None:
pep.log_metadata("taa", row[args.taa].upper() == "TAA")
expression = {k: max(v) for k, v in expression.iteritems()}
df_result = EpitopePredictionResult.from_dict(df_pred)
df_result.index = pandas.MultiIndex.from_tuples(
[tuple((i, "custom")) for i in df_result.index],
names=['Seq', 'Method'])
return df_result, distance, expression, uncertainty, pep_to_mutation
def toplevel_predictor(x):
predictor = EpitopePredictorFactory("netMHC", version="3.4")
peps = [Peptide(i) for i in x]
return predictor.predict(peps)
def setUp(self):
self.peptides = [
Peptide("KLLPRLPGV"),
Peptide("YLYDHLAPM"),
Peptide("ALYDVVSTL")
]
def solve(self, start=0, threads=None, options=None):
"""
Solve the epitope assembly problem with spacers optimally using integer linear programming.
.. note::
This can take quite long and should not be done for more and 30 epitopes max!
Also, one has to disable pre-solving steps in order to use this model.
:param int start: Start length for spacers (default 0).
:param int threads: Number of threads used for spacer design.
Be careful, if options contain solver threads it will allocate threads*solver_threads cores!
:param dict(str,str) options: Solver specific options as keys and parameters as values
:return: A list of ordered :class:`~Fred2.Core.Peptide.Peptide`
:rtype: list(:class:`~Fred2.Core.Peptide.Peptide`)
"""
def __load_model(name, model):
return getattr(
__import__("Fred2.Data.pssms." + name + ".mat." + model,
fromlist=[model]), model)
options = dict() if options is None else options
threads = mp.cpu_count() if threads is None else threads
pool = mp.Pool(threads)
#prepare parameters
cn = min(self.__clev_pred.supportedLength)
cl_pssm = __load_model(self.__clev_pred.name,
self.__clev_pred.name + "_" + str(cn))
cleav_pos = self.__clev_pred.cleavagePos
en = self.__en
epi_pssms = {}
allele_prob = {}
delete_alleles = []
if self.__epi_pred.name in ["smm", "smmpmbec", "comblibsidney"]:
self.__thresh = {
k: (1 - math.log(v, 50000) if v != 0 else 0)
for k, v in self.__thresh.items()
}
for a in self.__alleles:
allele_prob[a.name] = a.prob
try:
pssm = __load_model(
self.__epi_pred.name,
"%s_%i" % (self.__epi_pred.convert_alleles([a])[0], en))
if self.__epi_pred.name in [
"smm", "smmpmbec", "comblibsidney"
]:
for j, v in pssm.items():
for aa, score in v.items():
epi_pssms[j, aa, a.name] = 1 / 10. - math.log(
math.pow(10, score), 50000)
else:
for j, v in pssm.items():
for aa, score in v.items():
epi_pssms[j, aa, a.name] = score
except ImportError:
delete_alleles.append(a)
#delete alleles from model that generated an error while loading matrices
for a in delete_alleles:
del allele_prob[a.name]
del self.__thresh[a.name]
if not epi_pssms:
raise ValueError(
"Selected alleles with epitope length are not supported by the prediction method."
)
#print "run spacer designs in parallel using multiprocessing"
res = pool.map(
_runs_lexmin,
((str(ei), str(ej), i, en, cn, cl_pssm, epi_pssms, cleav_pos,
allele_prob, self.__alpha, self.__thresh, self.__solver,
self.__beta, options) for i in range(start, self.__k + 1)
for ei, ej in itr.product(self.__peptides, repeat=2) if ei != ej))
pool.close()
pool.join()
opt_spacer = {}
adj_matrix = {}
inf = float("inf")
#print res
#print "find best scoring spacer for each epitope pair"
for ei, ej, score, epi, spacer, c1, c2, non_c in res:
#print ei,spacer,ej,min(c1,c2),c1,c2
if adj_matrix.get((ei, ej), inf) > -min(c1, c2):
adj_matrix[(ei, ej)] = -min(c1, c2)
opt_spacer[(ei, ej)] = spacer
self.spacer = opt_spacer
#print "solve assembly with generated adjacency matrix"
assembler = EpitopeAssembly(self.__peptides,
self.__clev_pred,
solver=self.__solver,
matrix=adj_matrix)
res = assembler.solve(options=options)
#generate output
sob = []
for i in range(len(res) - 1):
ei = str(res[i])
ej = str(res[i + 1])
if not i:
sob.append(Peptide(ei))
sob.append(Peptide(opt_spacer[ei, ej]))
sob.append(Peptide(ej))
return sob
def __init__(self,
peptides,
pred,
solver="glpk",
weight=0.0,
matrix=None,
verbosity=0):
if not isinstance(pred, ACleavageSitePrediction):
raise ValueError(
"Cleave site predictor must be of type ACleavageSitePrediction"
)
if len(peptides) > 60:
warnings.warn(
"The peptide set exceeds 60. Above this level one has to expect "
+
"considerably long running times due to the complexity of the problem."
)
#Generate model
#1. Generate peptides for which cleave sites have to be predicted
#2. generate graph with dummy element
self.__verbosity = verbosity
pep_tmp = peptides[:]
pep_tmp.append("Dummy")
edge_matrix = {}
fragments = {}
seq_to_pep = {}
self.neo_cleavage = {}
self.good_cleavage = {}
if matrix is None:
for start, stop in itr.combinations(pep_tmp, 2):
if start == "Dummy" or stop == "Dummy":
seq_to_pep[str(start)] = start
seq_to_pep[str(stop)] = stop
edge_matrix[(str(start), str(stop))] = 0
edge_matrix[(str(stop), str(start))] = 0
else:
start_str = str(start)
stop_str = str(stop)
frag = Peptide(start_str + stop_str)
garf = Peptide(stop_str + start_str)
fragments[frag] = (start_str, stop_str)
fragments[garf] = (stop_str, start_str)
cleave_pred = pred.predict(list(fragments.keys()))
#cleave_site_df = cleave_pred.xs((slice(None), (cleavage_pos-1)))
for i in set(cleave_pred.index.get_level_values(0)):
fragment = "".join(cleave_pred.ix[i]["Seq"])
start, stop = fragments[fragment]
cleav_pos = len(str(start)) - 1
edge_matrix[(start, stop)] = -1.0 * (
cleave_pred.loc[(i, len(str(start)) - 1), pred.name] -
weight * sum(cleave_pred.loc[(i, j), pred.name]
for j in range(cleav_pos - 1, cleav_pos +
4, 1) if j != cleav_pos))
self.neo_cleavage[(start, stop)] = sum(
cleave_pred.loc[(i, j), pred.name]
for j in range(cleav_pos - 1, cleav_pos + 4, 1)
if j != cleav_pos)
self.good_cleavage[(start,
stop)] = cleave_pred.loc[(i,
len(str(start)) -
1), pred.name]
else:
edge_matrix = matrix
seq_to_pep = {str(p): p for p in pep_tmp}
for p in seq_to_pep.keys():
if p != "Dummy":
edge_matrix[(p, "Dummy")] = 0
edge_matrix[("Dummy", p)] = 0
self.__seq_to_pep = seq_to_pep
#3. initialize ILP
self.__solver = SolverFactory(solver)
model = ConcreteModel()
E = [x for x in list(seq_to_pep.keys()) if x != "Dummy"]
model.E = Set(initialize=E)
model.E_prime = Set(initialize=list(seq_to_pep.keys()))
model.ExE = Set(initialize=itr.permutations(E, 2), dimen=2)
model.w_ab = Param(model.E_prime,
model.E_prime,
initialize=edge_matrix)
model.card = Param(initialize=len(model.E_prime))
model.x = Var(model.E_prime, model.E_prime, within=Binary)
model.u = Var(model.E, domain=PositiveIntegers, bounds=(2, model.card))
model.obj = Objective(
rule=lambda mode: sum(model.w_ab[a, b] * model.x[a, b]
for a in model.E_prime for b in model.E_prime
if a != b),
sense=minimize)
model.tour_constraint_1 = Constraint(
model.E_prime,
rule=lambda model, a: sum(model.x[a, b] for b in model.E_prime
if a != b) == 1)
model.tour_constraint_2 = Constraint(
model.E_prime,
rule=lambda model, a: sum(model.x[b, a] for b in model.E_prime
if a != b) == 1)
model.cardinality_constraint = Constraint(
model.ExE,
rule=lambda model, a, b: model.u[a] - model.u[b] + 1 <=
(model.card - 1) * (1 - model.x[a, b]))
self.instance = model
if self.__verbosity > 0:
print("MODEL INSTANCE")
self.instance.pprint()
def approximate(self, start=0, threads=1, options=None):
"""
Approximates the Eptiope Assembly problem by applying Lin-Kernighan traveling salesman heuristic
LKH implementation must be downloaded, compiled, and globally executable.
Source code can be found here:
http://www.akira.ruc.dk/~keld/research/LKH/
:param int start: Start length for spacers (default 0).
:param int threads: Number of threads used for spacer design. Be careful, if options contain solver threads it
will allocate threads*solver_threads cores!
:param dict(str,str) options: Solver specific options (threads for example)
:return: A list of ordered :class:`~Fred2.Core.Peptide.Peptide`
:rtype: list(:class:`~Fred2.Core.Peptide.Peptide`)
"""
def __load_model(name, model):
return getattr(
__import__("Fred2.Data.pssms." + name + ".mat." + model,
fromlist=[model]), model)
options = dict() if options is None else options
threads = mp.cpu_count() if threads is None else threads
pool = mp.Pool(threads)
# prepare parameters
cn = min(self.__clev_pred.supportedLength)
cl_pssm = __load_model(self.__clev_pred.name,
self.__clev_pred.name + "_" + str(cn))
cleav_pos = self.__clev_pred.cleavagePos
en = self.__en
epi_pssms = {}
allele_prob = {}
delete_alleles = []
if self.__epi_pred.name in ["smm", "smmpmbec", "comblibsidney"]:
self.__thresh = {
k: (1 - math.log(v, 50000) if v != 0 else 0)
for k, v in self.__thresh.items()
}
for a in self.__alleles:
allele_prob[a.name] = a.prob
try:
pssm = __load_model(
self.__epi_pred.name,
"%s_%i" % (self.__epi_pred.convert_alleles([a])[0], en))
if self.__epi_pred.name in [
"smm", "smmpmbec", "comblibsidney"
]:
for j, v in pssm.items():
for aa, score in v.items():
epi_pssms[j, aa, a.name] = 1 / 10. - math.log(
math.pow(10, score), 50000)
else:
for j, v in pssm.items():
for aa, score in v.items():
epi_pssms[j, aa, a.name] = score
except ImportError:
delete_alleles.append(a)
# delete alleles from model that generated an error while loading matrices
for a in delete_alleles:
del allele_prob[a.name]
del self.__thresh[a.name]
if not epi_pssms:
raise ValueError(
"Selected alleles with epitope length are not supported by the prediction method."
)
# print "run spacer designs in parallel using multiprocessing"
res = pool.map(
_runs_lexmin,
((str(ei), str(ej), i, en, cn, cl_pssm, epi_pssms, cleav_pos,
allele_prob, self.__alpha, self.__thresh, self.__solver,
self.__beta, options) for i in range(start, self.__k + 1)
for ei, ej in itr.product(self.__peptides, repeat=2) if ei != ej))
pool.close()
pool.join()
opt_spacer = {}
adj_matrix = {}
inf = float("inf")
# print res
# print "find best scoring spacer for each epitope pair"
for ei, ej, score, epi, spacer, c1, c2, non_c in res:
if adj_matrix.get((ei, ej), inf) > -min(c1, c2):
adj_matrix[(ei, ej)] = -min(c1, c2)
opt_spacer[(ei, ej)] = spacer
self.spacer = opt_spacer
#print "solve assembly with generated adjacency matrix"
assembler = EpitopeAssembly(self.__peptides,
self.__clev_pred,
solver=self.__solver,
matrix=adj_matrix)
res = assembler.approximate()
#generate output
sob = []
for i in range(len(res) - 1):
ei = str(res[i])
ej = str(res[i + 1])
if not i:
sob.append(Peptide(ei))
sob.append(Peptide(opt_spacer[ei, ej]))
sob.append(Peptide(ej))
return sob
import pandas as pd
from docopt import docopt
if __name__ == "__main__":
arguments = docopt(__doc__)
file_in = arguments["--input"]
if not file_in:
file_in = "./data/binders.csv"
file_out = arguments["--output"]
dt = pd.read_csv(file_in)
dt = dt[dt["Sequence"].notnull()]
dt = dt[dt["Sequence"].str.len() == 9]
peptides = [Peptide(peptide) for peptide in dt["Sequence"]]
dt["allele"] = dt["allele"].str.replace("\*","").\
str.replace("(-[a-zA-Z]+)([0-9]{2})([0-9]{2})","\\1*\\2:\\3").\
str.replace("w","").\
str.replace("HLA-","")
dt.rename(columns={"Sequence": "peptide"}, inplace=True)
alleles = [Allele(allele) for allele in dt["allele"].unique().tolist()]
res = fred2wrap.predict_peptide_effects(
peptides, alleles=dt["allele"].unique().tolist())
res["peptide"] = [peptide.tostring() for peptide in res["peptide"]]
res["allele"] = [str(allele) for allele in res["allele"]]
res = res.pivot_table(index=["peptide", "allele"],
columns='method',
def main():
parser = argparse.ArgumentParser(
description=
"""The software is a novel approach to construct epitope-based string-of-beads
vaccines in optimal order and with sequence-optimized spacers of flexible length
such that the recovery of contained epitopes is maximized and immunogenicity of
arising neo-epitopes is reduced. """)
parser.add_argument("-i",
"--input",
required=True,
help="File containing epitopes (one peptide per line)")
parser.add_argument(
"-a",
"--alleles",
required=True,
help=
"Specifies file containing HLA alleles with corresponding HLA probabilities (one HLA per line)"
)
#parameters of the model
parser.add_argument(
"-k",
"--max_length",
default=6,
type=int,
help="Specifies the max. length of the spacers (default 6)")
parser.add_argument(
"-al",
"--alpha",
default=0.99,
type=float,
help=
"Specifies the first-order preference of the user in the model [0,1] (default 0.99)"
)
parser.add_argument(
"-be",
"--beta",
default=0.0,
type=float,
help=
"Specifies the second-order preference of the user in the model [0,1] (default 0)."
)
parser.add_argument(
"-cp",
"--cleavage_prediction",
default="PCM",
help=
"Specifies the used cleavage prediction method (default PCM) [available: PCM, PROTEASMM_C, PROTEASMM_S]"
)
parser.add_argument(
"-ep",
"--epitope_prediction",
default="Syfpeithi",
help=
"Specifies the used epitope prediction method (default Syfpeithi) [available: Syfpeithi, BIMAS, SMM, SMMPMBEC]"
)
parser.add_argument(
"-thr",
"--threshold",
default=20,
type=float,
help=
"Specifies epitope prediction threshold for SYFPEITHI (default 20).")
parser.add_argument("-o",
"--output",
required=True,
help="Specifies the output file.")
parser.add_argument(
"-t",
"--threads",
type=int,
default=None,
help=
"Specifies number of threads. If not specified all available logical cpus are used."
)
parser.add_argument(
"--ips-solver",
default="cplex",
choices=["cplex", "cbc"],
help=
"Executable name of the IPS solver. Executable needs to be available in PATH."
)
parser.add_argument("--tsp-solution",
default="approximate",
choices=["approximate", "optimal"],
help="Type of solution of the TSP")
parser.add_argument(
"--random-order",
action="store_true",
help=
"Indicate whether to generate a random ordered string-of-beads polypeptide"
)
parser.add_argument(
"--seed",
type=int,
default=1,
help="Seed for random ordering of string-of-beads polypeptide")
args = parser.parse_args()
#parse input
peptides = list(FileReader.read_lines(args.input, in_type=Peptide))
#read in alleles
alleles = generate_alleles(args.alleles)
if args.cleavage_prediction.upper() not in [
"PCM", "PROTEASMM_C", "PROTEASMM_S"
]:
print "Specified cleavage predictor is currently not supported. Please choose either PCM, PROTEASMM_C, or PROTEASMM_S"
sys.exit(-1)
if args.epitope_prediction.upper() not in [
"SYFPEITHI", "BIMAS", "SMM", "SMMPMBEC"
]:
print "Specified cleavage predictor is currently not supported. Please choose either Syfpeithi, BIMAS, SMM, SMMPMBEC"
sys.exit(-1)
#set-up model
cl_pred = CleavageSitePredictorFactory(args.cleavage_prediction)
epi_pred = EpitopePredictorFactory(args.epitope_prediction)
thr = {a.name: args.threshold for a in alleles}
solver = EpitopeAssemblyWithSpacer(peptides,
cl_pred,
epi_pred,
alleles,
k=args.max_length,
en=9,
threshold=thr,
solver=args.ips_solver,
alpha=args.alpha,
beta=args.beta,
verbosity=0)
#solve
#pre-processing has to be disable otherwise many solver will destroy the symmetry of the problem
#how to do this is dependent on the solver used. For CPLEX it is preprocessing_presolve=n
threads = mp.cpu_count() if args.threads is None else args.threads
if args.tsp_solution == "approximate":
svbws = solver.approximate(threads=threads,
options={
"preprocessing_presolve": "n",
"threads": 1
})
else:
svbws = solver.solve(threads=threads,
options={
"preprocessing_presolve": "n",
"threads": 1
})
# Generate random ordered string-of-breads, but still uses optimal spacers
# determined from the above solve function.
if args.random_order:
print "Generating a randomly ordered polypeptide"
random.seed(args.seed)
random_order_sob = []
random.shuffle(peptides)
for i in range(len(peptides)):
# Break from loop once we hit the last peptide
if i == len(peptides) - 1:
random_order_sob.extend([Peptide(str(peptides[i]))])
break
left_peptide = str(peptides[i])
right_peptide = str(peptides[i + 1])
opt_spacer = solver.spacer[(left_peptide, right_peptide)]
# Right peptide gets added in the next iteration
random_order_sob.extend(
[Peptide(left_peptide),
Peptide(opt_spacer)])
svbws = random_order_sob
print
print "Resulting String-of-Beads: ", "-".join(map(str, svbws))
print
with open(args.output, "w") as f:
f.write("-".join(map(str, svbws)))
if __name__ == "__main__":
arguments = docopt(__doc__)
file_in = arguments["--input"]
if not file_in:
file_in = os.path.expanduser("data/immunogenic_SNVs-training_sets.csv")
file_out = arguments["--output"]
if not file_out:
file_out = os.path.expanduser("data/immunogenic_SNVs-model_data.csv")
dt = pd.read_csv(file_in)
dt = dt[dt["mutant_sequence"].notnull() & dt["wt_sequence"].notnull()]
# dt = dt[dt["Sequence"].str.len() == 9]
all_peptides = dt["mutant_sequence"].append(dt["wt_sequence"]).unique()
peptides = [Peptide(peptide) for peptide in all_peptides]
dt["allele"] = dt["allele"].str.replace("\*", "").\
str.replace(":", "").\
str.replace("(-[a-zA-Z]+)([0-9]{2})([0-9]{2})", "\\1*\\2:\\3").\
str.replace("w", "").\
str.replace("HLA-", "")
# TODO
# dt.rename(columns = {"Sequence": "peptide"}, inplace = True)
alleles = []
valid_alleles = []
for allele in dt["allele"].tolist():
try:
a = Allele(allele)
valid_alleles.append(True)
parser.add_argument('alleles',
metavar='allele',
nargs='+',
type=str,
help='HLA alleles to predict against.')
args = parser.parse_args()
####################################################################################################
from Fred2.Core import Allele, Peptide
from Fred2.EpitopePrediction import EpitopePredictorFactory
####################################################################################################
# Convert raw peptide sequences to Fred2.Core.Peptide objects
all_peptides = [Peptide(row.strip()) for row in args.peptides]
# Separate peptides by length
peptides_by_length = {}
for peptide in all_peptides:
if not len(peptide) in peptides_by_length:
peptides_by_length[len(peptide)] = []
peptides_by_length[len(peptide)].append(peptide)
# Convert raw allele strings to Fred2.Core.Allele objects
alleles = [Allele(allele) for allele in args.alleles]
# Instatiate predictor
predictor = EpitopePredictorFactory("Syfpeithi")
def read_epitope_input(args, alleles, exclude):
"""
reads in epitope files generated by NGSAnalyzer+ImmogenicityPredictor
Header NGSAnalyzer:
mutation - position of the mutation in the reference genome (currently hg19); format:
chromosome_position; the position is zero-based
gene - gene affected by the mutation
transcript - transcript affected by the mutation (UCSC known genes transcript ID)
transcript_expression - expression in RPKM/FPKM of the affected transcript
neopeptide - peptide resulting from the mutation in the given transcript
length_of_neopeptide - length of the neopeptide
HLA - HLA used for the binding prediction of the neopeptide
HLA_class1_binding_prediction - predicted binding affinity (currently rank score of IEDB consensus tool)
Header ImmunogenicityPredictor:
immunogenicity - predicted immunogenicity for specific HLA allele in column
distance - distance-to-self estimation specific for HLA allele in column
[uncertainty] - If immunopredictor can estimate prediction uncertainty
:param args: Input arguments
:param alleles: HLA alleles
:param exlude: excluded peptides
:return: df_epitope - EpitopePredictionResult
distance - dict(pep_string, float)
expression - dict(gene_id, float)
uncertainty - dict(pep_string, float)
pep_to_mutation - dict(pep_string, mutation_string)
"""
distance = {}
uncertainty = {}
expression = {}
seq_to_pep = {}
gene_to_prot = {}
hla_to_allele = {a.name:a for a in alleles}
df_pred = {a:{} for a in alleles}
pep_to_mutation = {}
with open(args.input, "rU") as f:
reader = csv.DictReader(f, delimiter='\t')
for row in reader:
seq = row["neopeptide"]
if (exclude is None or seq not in exclude) and len(seq) > 0:
if seq in seq_to_pep:
pep = seq_to_pep[seq]
else:
pep = Peptide(seq.upper())
seq_to_pep[seq] = pep
try:
allele = hla_to_allele[row["HLA"].replace("HLA-","")]
except:
logging.warning(
"HLA {allele} was not contained in the provided allele file. Please check your input.".format(
allele=row["HLA"]))
continue
gene = row["gene"]
if gene in gene_to_prot:
prot = gene_to_prot[gene]
else:
prot = Protein("", gene_id=gene, transcript_id=gene)
gene_to_prot[gene] = prot
pep.proteins[prot.transcript_id]=prot
pep.proteinPos[prot.transcript_id].append(0)
pep_to_mutation.setdefault(seq, []).append(row["mutation"])
expression.setdefault(gene,[]).append(float(row["transcript_expression"]))
if args.rank:
df_pred[allele][pep] = max(0., 1.
- float(row[args.immunogenicity])/100.0)
else:
df_pred[allele][pep] = max(0., 1.
- math.log(float(row[args.immunogenicity]),
50000)) if args.immunogenicity == "HLA_class1_binding_prediction" else float(
row[args.immunogenicity])
if args.distance is not None:
distance[(seq,allele.name)] = float(row[args.distance])
if args.uncertainty is not None:
uncertainty[(seq,allele.name)] = float(row[args.uncertainty])
if args.taa is not None:
pep.log_metadata("taa", row[args.taa].upper() == "TAA" )
expression = {k:max(v) for k,v in expression.iteritems()}
df_result = EpitopePredictionResult.from_dict(df_pred)
df_result.index = pandas.MultiIndex.from_tuples([tuple((i, "custom")) for i in df_result.index],
names=['Seq', 'Method'])
return df_result, distance, expression, uncertainty, pep_to_mutation
def setUp(self):
self.seqs = [Peptide("SYFPEISYFP"),
Protein("IHTIEPFYSIHTIEPFYSIHTIEPFYSIHTIEPFYSIHTIEPFYS", transcript_id="ID-01", gene_id="FOXP3")]
self.transcript = Transcript("")
""")
_imm, _non = set(), set()
if args.dataset == 'iedb.tcell':
_imm, _non = pepdata.iedb.tcell.load_classes(nrows=args.n)
elif args.dataset == 'iedb.mhc':
_imm, _non = pepdata.iedb.mhc.load_classes(nrows=args.n)
elif args.dataset == 'imma2':
_imm, _non = pepdata.imma2.load_classes()
elif args.dataset == 'stdin':
_imm = [line for line in stdin]
else:
print("available datasets: iedb.tcell, ideb.mhc, imma2")
exit()
imm = [Peptide(elem) for elem in _imm]
non = [Peptide(elem) for elem in _non]
#hla='HLA-A2|HLA-A\*02'
#df = ()
#
#if args.dataset == 'iedb.tcell':
# df = pepdata.iedb.tcell.load_dataframe(nrows=args.n)
#elif args.dataset == 'iedb.mhc':
# df = pepdata.iedb.mhc.load_dataframe(nrows=args.n)
#elif args.dataset == 'iedb.bcell':
# df = pepdata.iedb.bcell.load_dataframe(nrows=args.n)
#elif args.dataset == 'stdin':
# df = [ line for line in stdin ]
#else:
# print("available datasets: iedb.tcell, ideb.mhc, imma2")
