def run_all(gbfile, groupfile, transfile, tagFiles):
# does this filter out replicates that appear in every tag?
annodb, al, dna = analyze.read_genbank_annots(gbfile)
snps = analyze.read_tag_files(tagFiles)
gsd = analyze.map_snps(snps, al, dna)
# Count nonsyn vs. syn. sites for each gene
site_counts = genes_sites_dict(annodb)
# Count nonsyn vs. syn. snps for each gene
snp_counts = analyze.get_gene_na_ns(gsd)
genes = annodb.keys()
functional_groups = load_func_assoc(groupfile, transfile)
#binomial_tests(snp_counts, site_counts)
#main(site_counts, snp_counts, genes, functional_groups, test_func=binomial_test)
for name, test_func in [("Fisher test", fisher_test),
("Binomial Test", binomial_test)]:
print name
run_tests(site_counts,
snp_counts,
genes,
functional_groups,
test_func=test_func)
print
def hypergeom_cmd(gbfile, groupfile, transfile, tagFiles, N=30):
top_genes = phenoseq_top_genes(gbfile, tagFiles)
pathway_dict = load_func_assoc(groupfile, transfile)
top_genes_subset = [y for (x,y) in top_genes[:N]]
#print top_genes_subset
results = []
for name, genes_ in pathway_dict.items():
genes = genes_[1]
num_genes_int_top_list = len([g for g in genes if g in top_genes_subset])
if num_genes_int_top_list:
pval = p_value(len(genes), num_genes_int_top_list,
len(top_genes_subset))
if isnan(pval):
warnings.warn('ignoring invalid NaN pvalues...')
else:
results.append( (len(pathway_dict) * pval, name,
len(genes), genes))
results.sort()
for p, name, n, genes in results:
print ",".join(map(str, [p, name, n, " ".join(genes)]))
def run_all(gbfile, groupfile, transfile, tagFiles):
# does this filter out replicates that appear in every tag?
annodb, al, dna = analyze.read_genbank_annots(gbfile)
snps = analyze.read_tag_files(tagFiles)
gsd = analyze.map_snps(snps, al, dna)
# Count nonsyn vs. syn. sites for each gene
site_counts = genes_sites_dict(annodb)
# Count nonsyn vs. syn. snps for each gene
snp_counts = analyze.get_gene_na_ns(gsd)
genes = annodb.keys()
functional_groups = load_func_assoc(groupfile, transfile)
#binomial_tests(snp_counts, site_counts)
#main(site_counts, snp_counts, genes, functional_groups, test_func=binomial_test)
for name, test_func in [("Fisher test", fisher_test), ("Binomial Test", binomial_test)]:
print name
run_tests(site_counts, snp_counts, genes, functional_groups,
test_func=test_func)
print
def hypergeom_cmd(gbfile, groupfile, transfile, tagFiles, N=30):
top_genes = phenoseq_top_genes(gbfile, tagFiles)
pathway_dict = load_func_assoc(groupfile, transfile)
top_genes_subset = [y for (x, y) in top_genes[:N]]
#print top_genes_subset
results = []
for name, genes_ in pathway_dict.items():
genes = genes_[1]
num_genes_int_top_list = len(
[g for g in genes if g in top_genes_subset])
if num_genes_int_top_list:
pval = p_value(len(genes), num_genes_int_top_list,
len(top_genes_subset))
if isnan(pval):
warnings.warn('ignoring invalid NaN pvalues...')
else:
results.append(
(len(pathway_dict) * pval, name, len(genes), genes))
results.sort()
for p, name, n, genes in results:
print ",".join(map(str, [p, name, n, " ".join(genes)]))