print "done!"
##
# Calculation begins at this point.
results = {}
database = mysql.connector.Connect(**DB_INFO)
db = database.cursor()
db.execute("SELECT rsid FROM diseases.neandertal_snps;")
for nrsid in [row[0] for row in db.fetchall()]:
user_nsnp = user_snps.get(nrsid, None)
# We could not get this SNP directly: impute.
if user_nsnp is None:
user_nsnp = genotype_tools.impute_rsid_simple(user_snps, nrsid, population)
print "Imputed %s -> %s." % (user_nsnp.nearest_SNP, user_nsnp.rsid,)
snp_info = get_snp_info(user_nsnp)
# If we imputed, user_nsnp.rsid is the imputed rsid, and nearest_SNP is the
# rsid of the SNP we were imputing from (imputing for).
if user_nsnp.nearest_SNP is not None:
snp_info["imputed_from"] = user_nsnp.nearest_SNP
results[user_nsnp.rsid] = snp_info
nnumerator = sum(a["score"][0] for a in results.values())
ndenominator = sum(a["score"][1] for a in results.values())
# Name the output HTML file based on the input genome file"s name.
filename = os.path.splitext(os.path.split(user_genome_path)[1])[0]
out_file_name = "%s.html" % (filename,)
# For each EL-SNP, we get the value of that SNP from the provided genome
# and adjust the running EL and AL scores according to those values from the
# paper.
# Track all SNPs for output.
results = {}
for el_rsid in el_rsids:
# If we already have a value for this SNP - use it.
user_el_snp = user_snps.get("rs" + el_rsid, None)
# Impute if we have to...
if user_el_snp is None:
try:
user_el_snp = genotype_tools.impute_rsid_simple(
user_snps, "rs" + el_rsid, population
)
# TODO: workaround error with rs2042831 and CEU from Mikolaj Habryn.
except ValueError, e:
print "Error occurred imputing for rs%s: %s." % (el_rsid, e)
continue
# This SNP was imputed.
print "Imputed %s." % (user_el_snp.nearest_SNP,)
# Imputation returned None (this should never happen).
if user_el_snp is None:
print "Unable to impute value for rs%s: imputation returned None." % el_rsid
continue
