def run(self):
sqlrunner = SQL(database=os.path.join(self.root, "databases", self.database))
gRNA_db = sqlrunner.get_global_gRNA(mismatch=str(self.mismatch))
# This is a rate limiting step
if bool(self.gene_mask_dict['genes']):
query_data = self.get_targeted_data(dataframe=gRNA_db, gene_mask_dict=self.gene_mask_dict)
else:
query_data = gRNA_db
multifasta = sqlrunner.get_gene_multifasta()
gRNA_runner = RefineCripri(grna_dataframe=query_data,
strand=self.strand,
fasta_dataframe=multifasta,
cas9=self.cas9_organism,
offtarget_ids=sqlrunner.custom_sql("SELECT name, strand FROM global_offtarget"))
candidates, backup, dropped = gRNA_runner.cripr_interference()
candidates, backup, dropped = map(self.utils.annotate_dataframe,
[candidates, backup, dropped])
offtargets = sqlrunner.get_offtargets_by_mismatch(mismatch=self.mismatch)
offtargets.dropna(subset=['annotation'], inplace=True)
offtargets = offtargets[offtargets['strand'] != '+']
offtargets['annotation'] = offtargets['annotation'].apply(
lambda x: x.replace("_", "") if isinstance(x, str) else x)
offtargets = offtargets.query("gene != annotation")
offtargets.reset_index(drop=True, inplace=True)
offtarget_ids = list(set(offtargets['name']))
candidates_has_offtargets = self.list_comparison(list1=candidates['names'], list2=offtarget_ids)
backup_has_offtargets = self.list_comparison(list1=backup['names'], list2=offtarget_ids)
if candidates_has_offtargets:
candidate_off_ids = list(set(candidates['names']) & set(offtarget_ids))
candidates_offtargets = self.grab_offtargets(query=candidates,
offtargets=offtargets,
offtarget_ids=offtarget_ids)
candidates = self.negate_pam_mismatch(grna_dataframe=candidates,
offtarget_dataframe=candidates_offtargets,
target_ids=candidate_off_ids)
candidates, dropped = self.move_grna_by_offtargets(grna_dataframe=candidates,
dropped_dataframe=dropped,
offtarget_dataframe=candidates_offtargets,
masks=self.gene_mask_dict['masks'])
candidates_offtargets = pd.DataFrame(candidates_offtargets)
else:
candidates_offtargets = dict.fromkeys(offtargets, [])
candidates_offtargets = pd.DataFrame(candidates_offtargets)
if backup_has_offtargets:
backup_off_ids = list(set(backup['names']) & set(offtarget_ids))
backup_offtargets = self.grab_offtargets(query=backup, offtargets=offtargets, offtarget_ids=offtarget_ids)
backup = self.negate_pam_mismatch(grna_dataframe=backup,
offtarget_dataframe=backup_offtargets,
target_ids=backup_off_ids)
backup, dropped = self.move_grna_by_offtargets(grna_dataframe=backup,
dropped_dataframe=dropped,
offtarget_dataframe=backup_offtargets,
masks=self.gene_mask_dict['masks'])
backup_offtargets = pd.DataFrame(backup_offtargets)
else:
backup_offtargets = dict.fromkeys(offtargets, [])
backup_offtargets = pd.DataFrame(backup_offtargets)
## add ranking to pam, move between dataframes if ranking is f****d
candidates, backup = self.scan_maxmismatches(candidates=candidates, backup=backup)
candidates, backup = self.force_max_grna_in_candidates(candidates=candidates, backup=backup,
max_grna=self.max_grna)
candidates = self.force_ag_base(dataframe=candidates, max_primer_size=self.max_primer_size)
backup = self.force_ag_base(dataframe=backup, max_primer_size=self.max_primer_size)
candidates, backup, dropped = map(self.calculate_primer_len, [candidates, backup, dropped])
candidates, backup, dropped = map(self.calculate_gc_content, [candidates, backup, dropped])
candidates = self.design_primers(dataframe=candidates, cas9=self.cas9_organism,
fiveprime=self.fiveprime, threeprime=self.threeprime)
backup = self.design_primers(dataframe=backup, cas9=self.cas9_organism,
fiveprime=self.fiveprime, threeprime=self.threeprime)
candidates, backup, dropped = map(pd.DataFrame,
[candidates, backup, dropped])
offtarget_empty = [candidates_offtargets.empty, backup_offtargets.empty]
final_offtargets = pd.DataFrame()
if not all(offtarget_empty):
final_offtargets = candidates_offtargets
final_offtargets['from'] = "candidates"
backup_offtargets['from'] = "backup"
final_offtargets = final_offtargets.append(backup_offtargets, ignore_index=True)
else:
if not offtarget_empty[0]:
final_offtargets = candidates_offtargets
final_offtargets['from'] = "candidates"
if not offtarget_empty[1]:
final_offtargets = backup_offtargets
final_offtargets['from'] = "backup"
if final_offtargets.empty:
final_offtargets = pd.DataFrame(columns=offtargets.columns)
candidates.to_csv(os.path.join(self.root, "temp", "candidates.txt"), header=True, index=False, sep=",")
backup.to_csv(os.path.join(self.root, "temp", "backup.txt"), header=True, index=False, sep=",")
dropped.to_csv(os.path.join(self.root, "temp", "dropped.txt"), header=True, index=False, sep=",")
final_offtargets.to_csv(os.path.join(self.root, "temp", "offtargets.txt"), header=True, index=False, sep=",")
class CrisprFuncHelpers:
"""
unittest class for the crispr.py
"""
def __init__(self, database: str, strand: str, mismatch: int, cas9: str):
self.root = os.path.dirname(os.path.abspath("../main.py"))
os.chdir(self.root)
self.sql = SQL(database=database)
self.strand = strand
self.mismatch = mismatch
self.cas9 = cas9
def initial_filter_test(self):
data = self.sql.get_global_gRNA(mismatch=self.mismatch)
genes = [genes.split("_")[0] for genes in data['names']]
data['genes'] = genes
query = ["Rv0899", "Rv0934"]
out = pd.DataFrame()
for items in query:
if items in genes:
grad_idx = [
idx for idx, val in data.iterrows()
if items in val['genes']
]
out = out.append(data.loc[grad_idx, :], ignore_index=True)
runner = RefineCripri(grna_dataframe=out,
strand=self.strand,
fasta_dataframe=None,
cas9=self.cas9)
candidates, backup, dropped = map(pd.DataFrame,
*[runner.initial_filter()])
candidates_out = list(
set([
True if row['score'] < 2 else False
for _, row in candidates.iterrows()
]))
backup_out = list(
set([
True if row['score'] >= 2 else False
for _, row in backup.iterrows()
]))
dropped_out = list(
set([
True if row['names'][-1] != self.strand else False
for _, row in dropped.iterrows()
]))
return [candidates_out[0], backup_out[0], dropped_out[0]]
def initial_filter_result(self):
return [True, True, True]
def has_offtarget_test(self):
data = self.sql.get_global_gRNA(mismatch=self.mismatch)
genes = [genes.split("_")[0] for genes in data['names']]
data['genes'] = genes
query = ["Rv0899", "Rv0934", "Rv0051"]
out = pd.DataFrame()
for items in query:
if items in genes:
grad_idx = [
idx for idx, val in data.iterrows()
if items in val['genes']
]
out = out.append(data.loc[grad_idx, :], ignore_index=True)
runner = RefineCripri(grna_dataframe=out,
strand=self.strand,
fasta_dataframe=None,
cas9=self.cas9)
candidates, backup, dropped = runner.initial_filter()
candidates, backup, dropped = runner.has_offtarget(
candidates=candidates, backup=backup, dropped_gRNA=dropped)
candidates, backup, dropped = map(pd.DataFrame,
[candidates, backup, dropped])
return True
def has_offtarget_result(self):
return True