Test whether MYC amplified samples correlate with TP53 mutations

In [5]:

    

import NotebookImport
from TCGA_analysis_PanCancer_import import *
from IPython.display import HTML


    

In [2]:

    

# 1) test whether MYC focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if not sample.checkFocalGeneAmp("MYC") and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if not sample.checkFocalGeneAmp("MYC") and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 1) test whether MYC focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors"
print "   MYC amplified samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC not amplified samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC amplified samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC not amplified samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 1) test whether MYC focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   MYC amplified samples having TP53 mutations: 81
   MYC not amplified samples having TP53 mutations: 1209
   MYC amplified samples having no TP53 mutation: 38
   MYC not amplified samples having no TP53 mutations: 1951
   Fisher-exact test: Odds-ratio: 3.43979365287   P-value: 1.63105139639e-10
   Fisher-exact test (one-sided greater): Odds-ratio: 3.43979365287   P-value: 1.03294439614e-10

In [3]:

    

# 1) Test whether MYC focally amplified non-breast tumors contain TP53 mutations more frequently than non focally amplified Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  if not sample.clinical:
     continue
  if sample.primarysiteofdesease == "breast":
     continue            
  if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if not sample.checkFocalGeneAmp("MYC") and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if not sample.checkFocalGeneAmp("MYC") and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 1) Test whether MYC focally amplified non-breast tumors contain TP53 mutations more frequently than non focally amplified Tumors"
print "   MYC amplified samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC not amplified samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC amplified samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC not amplified samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 1) Test whether MYC focally amplified non-breast tumors contain TP53 mutations more frequently than non focally amplified Tumors
   MYC amplified samples having TP53 mutations: 52
   MYC not amplified samples having TP53 mutations: 947
   MYC amplified samples having no TP53 mutation: 16
   MYC not amplified samples having no TP53 mutations: 1302
   Fisher-exact test: Odds-ratio: 4.46832101373   P-value: 1.66724769305e-08
   Fisher-exact test (one-sided greater): Odds-ratio: 4.46832101373   P-value: 1.43914292831e-08

In [3]:

    

# 2) test whether MYC and CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   MYC_amp_tp53_count=0
   MYC_no_amp_tp53_count=0
   MYC_amp_no_tp53_count=0
   MYC_no_amp_no_tp53_count=0

   for sample in samples.values():
      if not sample.somatic_mutation_data:
         continue
      if sample.checkFocalGeneAmp("MYC") and sample.checkFocalGeneAmp("CCND1") and "TP53" in sample.genes_affected:
         MYC_amp_tp53_count += 1
      if sample.checkFocalGeneAmp("MYC") and sample.checkFocalGeneAmp("CCND1") and "TP53" not in sample.genes_affected:
         MYC_amp_no_tp53_count += 1
      if not sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" in sample.genes_affected:
         MYC_no_amp_tp53_count += 1
      if not sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and "TP53" not in sample.genes_affected:
         MYC_no_amp_no_tp53_count += 1
   oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
   oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

   print ""
   print " 2) test whether MYC and CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors"
   print "   MYC and CCND1 amplified samples having TP53 mutations: "+str(MYC_amp_tp53_count)
   print "   MYC and CCND1 not amplified samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
   print "   MYC and CCND1 amplified samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
   print "   MYC and CCND1 not amplified samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
   print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
   print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 2) test whether MYC and CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   MYC and CCND1 amplified samples having TP53 mutations: 15
   MYC and CCND1 not amplified samples having TP53 mutations: 1064
   MYC and CCND1 amplified samples having no TP53 mutation: 3
   MYC and CCND1 not amplified samples having no TP53 mutations: 1799
   Fisher-exact test: Odds-ratio: 8.45394736842   P-value: 8.53448576145e-05
   Fisher-exact test (one-sided greater): Odds-ratio: 8.45394736842   P-value: 8.53448576148e-05

In [4]:

    

# 3) test whether CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   CCND1_amp_tp53_count=0
   CCND1_no_amp_tp53_count=0
   CCND1_amp_no_tp53_count=0
   CCND1_no_amp_no_tp53_count=0

   for sample in samples.values():
      if not sample.somatic_mutation_data:
         continue
      if sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and "TP53" in sample.genes_affected:
         CCND1_amp_tp53_count += 1
      if sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and "TP53" not in sample.genes_affected:
         CCND1_amp_no_tp53_count += 1
      if not sample.checkFocalGeneAmp("CCND1") and "TP53" in sample.genes_affected:
         CCND1_no_amp_tp53_count += 1
      if not sample.checkFocalGeneAmp("CCND1") and "TP53" not in sample.genes_affected:
         CCND1_no_amp_no_tp53_count += 1
   oddsratio, pvalue = scipy.stats.fisher_exact([[CCND1_amp_tp53_count, CCND1_amp_no_tp53_count], [CCND1_no_amp_tp53_count, CCND1_no_amp_no_tp53_count]])
   oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[CCND1_amp_tp53_count, CCND1_amp_no_tp53_count], [CCND1_no_amp_tp53_count, CCND1_no_amp_no_tp53_count]], alternative='greater')

   print ""
   print " 3) test whether CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors"
   print "   CCND1 amplified samples having TP53 mutations: "+str(CCND1_amp_tp53_count)
   print "   CCND1 not amplified samples having TP53 mutations: "+str(CCND1_no_amp_tp53_count)
   print "   CCND1 amplified samples having no TP53 mutation: "+str(CCND1_amp_no_tp53_count)
   print "   CCND1 not amplified samples having no TP53 mutations: "+str(CCND1_no_amp_no_tp53_count)
   print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
   print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 3) test whether CCND1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   CCND1 amplified samples having TP53 mutations: 145
   CCND1 not amplified samples having TP53 mutations: 1145
   CCND1 amplified samples having no TP53 mutation: 152
   CCND1 not amplified samples having no TP53 mutations: 1837
   Fisher-exact test: Odds-ratio: 1.53048149851   P-value: 0.000594109536398
   Fisher-exact test (one-sided greater): Odds-ratio: 1.53048149851   P-value: 0.00031902221439

In [5]:

    

# 4) test whether tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  copynumber = "n/a"
  log2 = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
  if (log2 != None):
     copynumber = 2.0 * 2.0 ** log2
  else:
     copynumber="n/a"
  if copynumber != "n/a" and copynumber > 4 and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if copynumber != "n/a" and copynumber < 4 and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if copynumber != "n/a" and copynumber > 4 and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if copynumber != "n/a" and copynumber < 4 and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 4) test whether tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors"
print "   MYC CN>4 samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC CN<4 samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC CN>4 samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC CN<4 samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 4) test whether tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
   MYC CN>4 samples having TP53 mutations: 150
   MYC CN<4 samples having TP53 mutations: 1152
   MYC CN>4 samples having no TP53 mutation: 61
   MYC CN<4 samples having no TP53 mutations: 1931
   Fisher-exact test (two-sided): Odds-ratio: 4.12184084699   P-value: 1.08960838322e-21
   Fisher-exact test (one-sided greater): Odds-ratio: 4.12184084699   P-value: 8.25964240337e-22

In [4]:

    

# 4) test whether non-breast tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  if not sample.clinical:
     continue
  if sample.primarysiteofdesease == "breast":
     continue   
  copynumber = "n/a"
  log2 = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
  if (log2 != None):
     copynumber = 2.0 * 2.0 ** log2
  else:
     copynumber="n/a"
  if copynumber != "n/a" and copynumber > 4 and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if copynumber != "n/a" and copynumber < 4 and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if copynumber != "n/a" and copynumber > 4 and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if copynumber != "n/a" and copynumber < 4 and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 4) test whether non-breast tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors"
print "   MYC CN>4 samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC CN<4 samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC CN>4 samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC CN<4 samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 4) test whether non-breast tumors with copynumber > 4 in MYC contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
   MYC CN>4 samples having TP53 mutations: 87
   MYC CN<4 samples having TP53 mutations: 918
   MYC CN>4 samples having no TP53 mutation: 13
   MYC CN<4 samples having no TP53 mutations: 1305
   Fisher-exact test (two-sided): Odds-ratio: 9.51357466063   P-value: 2.36719407203e-20
   Fisher-exact test (one-sided greater): Odds-ratio: 9.51357466063   P-value: 1.94752063172e-20

In [6]:

    

# 5) test whether tumors with copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with CCND1 CN<4 Tumors
CCND1_amp_tp53_count=0
CCND1_no_amp_tp53_count=0
CCND1_amp_no_tp53_count=0
CCND1_no_amp_no_tp53_count=0

for sample in samples.values():
    if not sample.somatic_mutation_data:
        continue
    copynumber = "n/a"
    log2 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
    if (log2 != None):
        copynumber = 2.0 * 2.0 ** log2
    else:
        copynumber="n/a"
    if copynumber != "n/a" and copynumber > 4 and "TP53" in sample.genes_affected:
        CCND1_amp_tp53_count += 1
    if copynumber != "n/a" and copynumber < 4 and "TP53" in sample.genes_affected:
        CCND1_no_amp_tp53_count += 1
    if copynumber != "n/a" and copynumber > 4 and "TP53" not in sample.genes_affected:
        CCND1_amp_no_tp53_count += 1
    if copynumber != "n/a" and copynumber < 4 and "TP53" not in sample.genes_affected:
        CCND1_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[CCND1_amp_tp53_count, CCND1_amp_no_tp53_count], [CCND1_no_amp_tp53_count, CCND1_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[CCND1_amp_tp53_count, CCND1_amp_no_tp53_count], [CCND1_no_amp_tp53_count, CCND1_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 5) test whether tumors with copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with CCND1 CN<4 Tumors"
print "   CCND1 CN>4 samples having TP53 mutations: "+str(CCND1_amp_tp53_count)
print "   CCND1 CN<4 samples having TP53 mutations: "+str(CCND1_no_amp_tp53_count)
print "   CCND1 CN>4 samples having no TP53 mutation: "+str(CCND1_amp_no_tp53_count)
print "   CCND1 CN<4 samples having no TP53 mutations: "+str(CCND1_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 5) test whether tumors with copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with CCND1 CN<4 Tumors
   CCND1 CN>4 samples having TP53 mutations: 123
   CCND1 CN<4 samples having TP53 mutations: 1180
   CCND1 CN>4 samples having no TP53 mutation: 104
   CCND1 CN<4 samples having no TP53 mutations: 1884
   Fisher-exact test (two-sided): Odds-ratio: 1.88829856584   P-value: 4.34084997995e-06
   Fisher-exact test (one-sided greater): Odds-ratio: 1.88829856584   P-value: 2.87655495243e-06

In [7]:

    

# 6) test whether tumors with copynumber > 4 in MYC but coynumber < 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  copynumber_MYC = "n/a"
  log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
  if (log2_MYC != None):
     copynumber_MYC = 2.0 * 2.0 ** log2_MYC
  else:
     copynumber_MYC="n/a"
  copynumber_CCND1 = "n/a"
  log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
  if (log2_CCND1 != None):
     copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
  else:
     copynumber_CCND1="n/a"
  if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC < 4 and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC < 4 and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 6) test whether tumors with copynumber > 4 in MYC but copynumber < 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors"
print "   MYC CN>4, CCND1 CN<4 samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC CN<4 samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC CN>4, CCND1 CN<4  samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC CN<4 samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 6) test whether tumors with copynumber > 4 in MYC but copynumber < 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
   MYC CN>4, CCND1 CN<4 samples having TP53 mutations: 132
   MYC CN<4 samples having TP53 mutations: 1152
   MYC CN>4, CCND1 CN<4  samples having no TP53 mutation: 52
   MYC CN<4 samples having no TP53 mutations: 1931
   Fisher-exact test (two-sided): Odds-ratio: 4.25500801282   P-value: 7.72649139447e-20
   Fisher-exact test (one-sided greater): Odds-ratio: 4.25500801282   P-value: 4.50285718666e-20

In [8]:

    

# 7) test whether tumors with copynumber > 4 in MYC and copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  copynumber_MYC = "n/a"
  log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
  if (log2_MYC != None):
     copynumber_MYC = 2.0 * 2.0 ** log2_MYC
  else:
     copynumber_MYC="n/a"
  copynumber_CCND1 = "n/a"
  log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
  if (log2_CCND1 != None):
     copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
  else:
     copynumber_CCND1="n/a"
  if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and "TP53" in sample.genes_affected:
     MYC_amp_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC < 4 and "TP53" in sample.genes_affected:
     MYC_no_amp_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and "TP53" not in sample.genes_affected:
     MYC_amp_no_tp53_count += 1
  if copynumber_MYC != "n/a" and copynumber_MYC < 4 and "TP53" not in sample.genes_affected:
     MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 7) test whether tumors with copynumber > 4 in MYC and copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors"
print "   MYC CN>4, CCND1 CN<4 samples having TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC CN<4 samples having TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC CN>4, CCND1 CN<4  samples having no TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC CN<4 samples having no TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 7) test whether tumors with copynumber > 4 in MYC and copynumber > 4 in CCND1 contain TP53 mutations more frequently than tumors with MYC CN<4 Tumors
   MYC CN>4, CCND1 CN<4 samples having TP53 mutations: 17
   MYC CN<4 samples having TP53 mutations: 1152
   MYC CN>4, CCND1 CN<4  samples having no TP53 mutation: 9
   MYC CN<4 samples having no TP53 mutations: 1931
   Fisher-exact test (two-sided): Odds-ratio: 3.16618441358   P-value: 0.0042412984064
   Fisher-exact test (one-sided greater): Odds-ratio: 3.16618441358   P-value: 0.00362973038831

In [9]:

    

%matplotlib inline


    

In [10]:

    

# 8) check TP53 mutation spectrum
TP53_types = list()

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      elif mutation.mut_type == "Silent":
         TP53_types.append("Silent")
      elif mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins":
         TP53_types.append("Frameshift_Indel")
      elif mutation.mut_type == "In_Frame_Del" or mutation.mut_type == "In_Frame_Ins":
         TP53_types.append("Inframe_Indel")
      elif mutation.mut_type == "Splice_Site":
         TP53_types.append("Splice-Site")
      elif mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7578264 and mutation.pos <= 7578443:
         TP53_types.append("L2_Missense")
      elif mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7577528 and mutation.pos <= 7577575:
         TP53_types.append("L3_Missense")
      elif mutation.mut_type == "Missense_Mutation":
         TP53_types.append("Other_Missense")
      elif mutation.mut_type == "Nonsense_Mutation":
         TP53_types.append("Nonsense")
      elif mutation.mut_type == "Nonstop_Mutation":
         TP53_types.append("Nonstop")
      elif mutation.mut_type == "RNA":
         TP53_types.append("RNA")

TP53_type_series = pandas.Categorical(sorted(TP53_types))
fig_TP53=TP53_type_series.describe().counts.plot(kind='pie', figsize=(6, 6),colors=['b', 'g', 'r', 'c', 'm', 'y','k','w'], title=None)
fig_TP53.set_ylabel('')

output1=fig_TP53.get_figure()
output1.savefig('PanCancer_mut_spectrum.png', dpi=200)

print ""
print " 8) check TP53 mutation spectrum"
print "   Total mutations: "+str(len(TP53_types))
print "   "
count = 0
s = "<table><tr><th>Category</th><th>Count</th><th>Frequency</th></tr>"
for i in TP53_type_series.categories:
   s += "<tr><td>"+i+"</td><td>"+str(TP53_type_series.describe()['counts'][count])+"</td><td>"+str(TP53_type_series.describe()['freqs'][count])+"</td></tr>"
   count += 1
s += "</table>"
h = HTML(s);h


    

    




 8) check TP53 mutation spectrum
   Total mutations: 1486
   






    
Out[10]:




Category
Count
Frequency
Frameshift_Indel
201
0.135262449529
Inframe_Indel
27
0.0181695827725
L2_Missense
150
0.100942126514
L3_Missense
143
0.0962314939435
Nonsense
196
0.131897711978
Other_Missense
663
0.446164199192
Silent
31
0.0208613728129
Splice-Site
75
0.0504710632571






    

In [11]:

    

# 9) check TP53 mutation spectrum in MYC amplified tumors
TP53_types_MYC = list()

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Silent":
         TP53_types_MYC.append("Silent")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         TP53_types_MYC.append("Frameshift_Indel")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and (mutation.mut_type == "In_Frame_Del" or mutation.mut_type == "In_Frame_Ins"):
         TP53_types_MYC.append("Inframe_Indel")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Splice_Site":
         TP53_types_MYC.append("Splice-Site")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7578264 and mutation.pos <= 7578443:
         TP53_types_MYC.append("L2_Missense")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7577528 and mutation.pos <= 7577575:
         TP53_types_MYC.append("L3_Missense")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Missense_Mutation":
         TP53_types_MYC.append("Other_Missense")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Nonsense_Mutation":
         TP53_types_MYC.append("Nonsense")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "Nonstop_Mutation":
         TP53_types_MYC.append("Nonstop")
      elif sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and mutation.mut_type == "RNA":
         TP53_types_MYC.append("RNA")

TP53_type_MYC_series = pandas.Categorical(sorted(TP53_types_MYC))
fig_TP53=TP53_type_MYC_series.describe().counts.plot(kind='pie', figsize=(6, 6),colors=['b', 'g', 'r', 'c', 'm', 'y','k','w'])
fig_TP53.set_ylabel('')

output1=fig_TP53.get_figure()
output1.savefig('PanCancer_MYC_focal_mut_spectrum.png', dpi=200)

print ""
print " 9) check TP53 mutation spectrum in MYC amplified tumors"
print "   Total mutations: "+str(len(TP53_types_MYC))
print "   "
count = 0
s = "<table><tr><th>Category</th><th>Count</th><th>Frequency</th></tr>"
for i in TP53_type_MYC_series.categories:
   s += "<tr><td>"+i+"</td><td>"+str(TP53_type_MYC_series.describe()['counts'][count])+"</td><td>"+str(TP53_type_MYC_series.describe()['freqs'][count])+"</td></tr>"
   count += 1
s += "</table>"
h = HTML(s);h


    

    




 9) check TP53 mutation spectrum in MYC amplified tumors
   Total mutations: 82
   






    
Out[11]:




Category
Count
Frequency
Frameshift_Indel
10
0.121951219512
Inframe_Indel
2
0.0243902439024
L2_Missense
11
0.134146341463
L3_Missense
7
0.0853658536585
Nonsense
14
0.170731707317
Other_Missense
32
0.390243902439
Silent
2
0.0243902439024
Splice-Site
4
0.0487804878049






    

In [33]:

    

# 10) Check mutation spectrum in MYC (and not CCND1) samples having copynumber > 4
TP53_types_MYC = list()

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   copynumber_MYC = "n/a"
   log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
   if (log2_MYC != None):
      copynumber_MYC = 2.0 * 2.0 ** log2_MYC
   else:
      copynumber_MYC="n/a"
   copynumber_CCND1 = "n/a"
   log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
   if (log2_CCND1 != None):
      copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
   else:
      copynumber_CCND1="n/a"
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Silent":
         TP53_types_MYC.append("Silent")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         TP53_types_MYC.append("Frameshift_Indel")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and (mutation.mut_type == "In_Frame_Del" or mutation.mut_type == "In_Frame_Ins"):
         TP53_types_MYC.append("Inframe_Indel")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Splice_Site":
         TP53_types_MYC.append("Splice-Site")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7578264 and mutation.pos <= 7578443:
         TP53_types_MYC.append("L2_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7577528 and mutation.pos <= 7577575:
         TP53_types_MYC.append("L3_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Missense_Mutation":
         TP53_types_MYC.append("Other_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Nonsense_Mutation":
         TP53_types_MYC.append("Nonsense")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "Nonstop_Mutation":
         TP53_types_MYC.append("Nonstop")
      elif copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and mutation.mut_type == "RNA":
         TP53_types_MYC.append("RNA")

TP53_type_MYC_series = pandas.Categorical(sorted(TP53_types_MYC))
fig_TP53=TP53_type_MYC_series.describe().counts.plot(kind='pie', figsize=(6, 6),colors=['b', 'g','r', 'c', 'm', 'y','k','w'])
fig_TP53.set_ylabel('')

output1=fig_TP53.get_figure()
output1.savefig('PanCancer_MYC_cn4_mut_spectrum.png', dpi=200)

print ""
print " 10) Check mutation spectrum in MYC (and not CCND1) samples having copynumber > 4"
print "   Total mutations: "+str(len(TP53_types_MYC))
print "   "
count = 0
s = "<table><tr><th>Category</th><th>Count</th><th>Frequency</th></tr>"
for i in TP53_type_MYC_series.categories:
   s += "<tr><td>"+i+"</td><td>"+str(TP53_type_MYC_series.describe()['counts'][count])+"</td><td>"+str(TP53_type_MYC_series.describe()['freqs'][count])+"</td></tr>"
   count += 1
s += "</table>"
h = HTML(s);h


    

    




 10) Check mutation spectrum in MYC (and not CCND1) samples having copynumber > 4
   Total mutations: 131
   






    
Out[33]:




Category
Count
Frequency
Frameshift_Indel
16
0.12213740458
Inframe_Indel
1
0.00763358778626
L2_Missense
10
0.0763358778626
L3_Missense
10
0.0763358778626
Nonsense
22
0.167938931298
Other_Missense
65
0.496183206107
Silent
2
0.0152671755725
Splice-Site
5
0.0381679389313






    

In [13]:

    

# 11) check TP53 mutation spectrum in CCND1 amplified tumors
TP53_types_CCND1 = list()

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Silent":
         TP53_types_CCND1.append("Silent")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         TP53_types_CCND1.append("Frameshift_Indel")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and (mutation.mut_type == "In_Frame_Del" or mutation.mut_type == "In_Frame_Ins"):
         TP53_types_CCND1.append("Inframe_Indel")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Splice_Site":
         TP53_types_CCND1.append("Splice-Site")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7578264 and mutation.pos <= 7578443:
         TP53_types_CCND1.append("L2_Missense")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7577528 and mutation.pos <= 7577575:
         TP53_types_CCND1.append("L3_Missense")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Missense_Mutation":
         TP53_types_CCND1.append("Other_Missense")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Nonsense_Mutation":
         TP53_types_CCND1.append("Nonsense")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "Nonstop_Mutation":
         TP53_types_CCND1.append("Nonstop")
      elif sample.checkFocalGeneAmp("CCND1") and not sample.checkFocalGeneAmp("MYC") and mutation.mut_type == "RNA":
         TP53_types_CCND1.append("RNA")

TP53_types_CCND1_series = pandas.Categorical(sorted(TP53_types_CCND1))
fig_TP53=TP53_types_CCND1_series.describe().counts.plot(kind='pie', figsize=(6, 6),colors=['b', 'g', 'r', 'c', 'm', 'y','k','w'])
fig_TP53.set_ylabel('')

output1=fig_TP53.get_figure()
output1.savefig('PanCancer_CCND1_focal_mut_spectrum.png', dpi=200)

print ""
print " 11) check TP53 mutation spectrum in CCND1 focally amplified tumors"
print "   Total mutations: "+str(len(TP53_types_CCND1))
print "   "
count = 0
s = "<table><tr><th>Category</th><th>Count</th><th>Frequency</th></tr>"
for i in TP53_types_CCND1_series.categories:
   s += "<tr><td>"+i+"</td><td>"+str(TP53_types_CCND1_series.describe()['counts'][count])+"</td><td>"+str(TP53_types_CCND1_series.describe()['freqs'][count])+"</td></tr>"
   count += 1
s += "</table>"
h = HTML(s);h


    

    




 11) check TP53 mutation spectrum in CCND1 focally amplified tumors
   Total mutations: 178
   






    
Out[13]:




Category
Count
Frequency
Frameshift_Indel
25
0.140449438202
Inframe_Indel
4
0.0224719101124
L2_Missense
24
0.134831460674
L3_Missense
18
0.101123595506
Nonsense
20
0.112359550562
Other_Missense
69
0.387640449438
Silent
5
0.0280898876404
Splice-Site
13
0.0730337078652






    

In [14]:

    

# 12) Check mutation spectrum in CCND1 (and not MYC) samples having copynumber > 4
TP53_types_CCND1 = list()

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   copynumber_MYC = "n/a"
   log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
   if (log2_MYC != None):
      copynumber_MYC = 2.0 * 2.0 ** log2_MYC
   else:
      copynumber_MYC="n/a"
   copynumber_CCND1 = "n/a"
   log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
   if (log2_CCND1 != None):
      copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
   else:
      copynumber_CCND1="n/a"
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Silent":
         TP53_types_CCND1.append("Silent")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         TP53_types_CCND1.append("Frameshift_Indel")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and (mutation.mut_type == "In_Frame_Del" or mutation.mut_type == "In_Frame_Ins"):
         TP53_types_CCND1.append("Inframe_Indel")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Splice_Site":
         TP53_types_CCND1.append("Splice-Site")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7578264 and mutation.pos <= 7578443:
         TP53_types_CCND1.append("L2_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Missense_Mutation" and mutation.pos >=  7577528 and mutation.pos <= 7577575:
         TP53_types_CCND1.append("L3_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Missense_Mutation":
         TP53_types_CCND1.append("Other_Missense")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Nonsense_Mutation":
         TP53_types_CCND1.append("Nonsense")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "Nonstop_Mutation":
         TP53_types_CCND1.append("Nonstop")
      elif copynumber_MYC != "n/a" and copynumber_MYC < 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 > 4 and mutation.mut_type == "RNA":
         TP53_types_CCND1.append("RNA")

TP53_types_CCND1_series = pandas.Categorical(sorted(TP53_types_CCND1))
fig_TP53=TP53_types_CCND1_series.describe().counts.plot(kind='pie', figsize=(6, 6),colors=['b', 'g','r', 'c', 'm', 'y','k','w'])
fig_TP53.set_ylabel('')

output1=fig_TP53.get_figure()
output1.savefig('PanCancer_CCND1_cn4_mut_spectrum.png', dpi=200)


print ""
print " 12) Check mutation spectrum in CCND1 (and not MYC) samples having copynumber > 4"
print "   Total mutations: "+str(len(TP53_types_CCND1))
print "   "
count = 0
s = "<table><tr><th>Category</th><th>Count</th><th>Frequency</th></tr>"
for i in TP53_types_CCND1_series.categories:
   s += "<tr><td>"+i+"</td><td>"+str(TP53_types_CCND1_series.describe()['counts'][count])+"</td><td>"+str(TP53_types_CCND1_series.describe()['freqs'][count])+"</td></tr>"
   count += 1
s += "</table>"
h = HTML(s);h


    

    




 12) Check mutation spectrum in CCND1 (and not MYC) samples having copynumber > 4
   Total mutations: 131
   






    
Out[14]:




Category
Count
Frequency
Frameshift_Indel
17
0.129770992366
Inframe_Indel
2
0.0152671755725
L2_Missense
21
0.160305343511
L3_Missense
13
0.0992366412214
Nonsense
17
0.129770992366
Other_Missense
47
0.358778625954
Silent
5
0.0381679389313
Splice-Site
9
0.0687022900763






    

In [15]:

    

# 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) focally amplified samples
MYC_amp_frameshift_count = 0
MYC_amp_no_frameshift_count = 0
MYC_no_amp_frameshift_count = 0
MYC_no_amp_no_frameshift_count = 0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         MYC_amp_frameshift_count += 1
      if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins":
         MYC_amp_no_frameshift_count += 1
      if not sample.checkFocalGeneAmp("MYC") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         MYC_no_amp_frameshift_count += 1
      if not sample.checkFocalGeneAmp("MYC") and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins":
         MYC_no_amp_no_frameshift_count += 1

oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]], alternative='greater')

print ""
print " 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) focally amplified samples"
print "   MYC amplified Frameshift Indels: "+str(MYC_amp_frameshift_count)
print "   MYC amplified other mutations: "+str(MYC_amp_no_frameshift_count)
print "   MYC not amplified Frameshift Indels: "+str(MYC_no_amp_frameshift_count)
print "   MYC not amplified other mutations: "+str(MYC_no_amp_no_frameshift_count)
print "   "
print "   Fisher exact test (two-sided): "
print "      -- odd's ratio: "+str(oddsratio)
print "      -- p-value: "+str(pvalue)
print "   Fisher exact test (one-sided greater): "
print "      -- odd's ratio: "+str(oddsratio_greater)
print "      -- p-value: "+str(pvalue_greater)


    

    




 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) focally amplified samples
   MYC amplified Frameshift Indels: 10
   MYC amplified other mutations: 78
   MYC not amplified Frameshift Indels: 187
   MYC not amplified other mutations: 1232
   
   Fisher exact test (two-sided): 
      -- odd's ratio: 0.844645550528
      -- p-value: 0.745117201673
   Fisher exact test (one-sided greater): 
      -- odd's ratio: 0.844645550528
      -- p-value: 0.736221318939

In [16]:

    

# 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) CN>4 samples
MYC_amp_frameshift_count = 0
MYC_amp_no_frameshift_count = 0
MYC_no_amp_frameshift_count = 0
MYC_no_amp_no_frameshift_count = 0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   copynumber_MYC = "n/a"
   log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
   if (log2_MYC != None):
      copynumber_MYC = 2.0 * 2.0 ** log2_MYC
   else:
      copynumber_MYC="n/a"
   copynumber_CCND1 = "n/a"
   log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
   if (log2_CCND1 != None):
      copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
   else:
      copynumber_CCND1="n/a"
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         MYC_amp_frameshift_count += 1
      if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins":
         MYC_amp_no_frameshift_count += 1
      if copynumber_MYC != "n/a" and copynumber_MYC < 4 and not sample.checkFocalGeneAmp("MYC") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins"):
         MYC_no_amp_frameshift_count += 1
      if copynumber_MYC != "n/a" and copynumber_MYC < 4 and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins":
         MYC_no_amp_no_frameshift_count += 1

oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]], alternative='greater')

print ""
print " 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) CN>4 samples"
print "   MYC amplified Frameshift Indels: "+str(MYC_amp_frameshift_count)
print "   MYC amplified other mutations: "+str(MYC_amp_no_frameshift_count)
print "   MYC not amplified Frameshift Indels: "+str(MYC_no_amp_frameshift_count)
print "   MYC not amplified other mutations: "+str(MYC_no_amp_no_frameshift_count)
print "   "
print "   Fisher exact test (two-sided): "
print "      -- odd's ratio: "+str(oddsratio)
print "      -- p-value: "+str(pvalue)
print "   Fisher exact test (one-sided greater): "
print "      -- odd's ratio: "+str(oddsratio_greater)
print "      -- p-value: "+str(pvalue_greater)


    

    




 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) CN>4 samples
   MYC amplified Frameshift Indels: 16
   MYC amplified other mutations: 122
   MYC not amplified Frameshift Indels: 173
   MYC not amplified other mutations: 1182
   
   Fisher exact test (two-sided): 
      -- odd's ratio: 0.896048517009
      -- p-value: 0.788639002871
   Fisher exact test (one-sided greater): 
      -- odd's ratio: 0.896048517009
      -- p-value: 0.694605165158

In [17]:

    

# 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) focally amplified samples
MYC_amp_frameshift_count = 0
MYC_amp_no_frameshift_count = 0
MYC_no_amp_frameshift_count = 0
MYC_no_amp_no_frameshift_count = 0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation" ):
         MYC_amp_frameshift_count += 1
      if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins" and not mutation.mut_type == "Nonsense_Mutation":
         MYC_amp_no_frameshift_count += 1
      if not sample.checkFocalGeneAmp("MYC") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation"):
         MYC_no_amp_frameshift_count += 1
      if not sample.checkFocalGeneAmp("MYC") and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins" and not mutation.mut_type == "Nonsense_Mutation":
         MYC_no_amp_no_frameshift_count += 1

oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]], alternative='greater')

print ""
print " 13) Test whether Deleterious mutations occur more frequently in MYC (and not CCND1) focally amplified samples"
print "   MYC amplified deleterious mutations: "+str(MYC_amp_frameshift_count)
print "   MYC amplified other mutations: "+str(MYC_amp_no_frameshift_count)
print "   MYC not amplified deleterious mutations: "+str(MYC_no_amp_frameshift_count)
print "   MYC not amplified other mutations: "+str(MYC_no_amp_no_frameshift_count)
print "   "
print "   Fisher exact test (two-sided): "
print "      -- odd's ratio: "+str(oddsratio)
print "      -- p-value: "+str(pvalue)
print "   Fisher exact test (one-sided greater): "
print "      -- odd's ratio: "+str(oddsratio_greater)
print "      -- p-value: "+str(pvalue_greater)


    

    




 13) Test whether Deleterious mutations occur more frequently in MYC (and not CCND1) focally amplified samples
   MYC amplified deleterious mutations: 24
   MYC amplified other mutations: 64
   MYC not amplified deleterious mutations: 368
   MYC not amplified other mutations: 1051
   
   Fisher exact test (two-sided): 
      -- odd's ratio: 1.07099184783
      -- p-value: 0.802452220641
   Fisher exact test (one-sided greater): 
      -- odd's ratio: 1.07099184783
      -- p-value: 0.432569742738

In [18]:

    

# 13) Test whether Frameshift indels occur more frequently in MYC (and not CCND1) CN>4
MYC_amp_frameshift_count = 0
MYC_amp_no_frameshift_count = 0
MYC_no_amp_frameshift_count = 0
MYC_no_amp_no_frameshift_count = 0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   copynumber_MYC = "n/a"
   log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
   if (log2_MYC != None):
      copynumber_MYC = 2.0 * 2.0 ** log2_MYC
   else:
      continue
   copynumber_CCND1 = "n/a"
   log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
   if (log2_CCND1 != None):
      copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
   else:
      continue
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if copynumber_MYC > 4 and copynumber_CCND1 < 4 and not sample.checkFocalGeneAmp("CCND1") and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation" ):
         MYC_amp_frameshift_count += 1
      if copynumber_MYC > 4 and copynumber_CCND1 < 4 and not sample.checkFocalGeneAmp("CCND1") and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins" and not mutation.mut_type == "Nonsense_Mutation":
         MYC_amp_no_frameshift_count += 1
      if copynumber_MYC < 4 and (mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation"):
         MYC_no_amp_frameshift_count += 1
      if copynumber_MYC < 4 and not mutation.mut_type == "Frame_Shift_Del" and not mutation.mut_type == "Frame_Shift_Ins" and not mutation.mut_type == "Nonsense_Mutation":
         MYC_no_amp_no_frameshift_count += 1

oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_frameshift_count, MYC_amp_no_frameshift_count], [MYC_no_amp_frameshift_count, MYC_no_amp_no_frameshift_count]], alternative='greater')

print ""
print " 13) Test whether Deleterious mutations occur more frequently in MYC (and not CCND1) CN>4 samples"
print "   MYC amplified deleterious mutations: "+str(MYC_amp_frameshift_count)
print "   MYC amplified other mutations: "+str(MYC_amp_no_frameshift_count)
print "   MYC not amplified deleterious mutations: "+str(MYC_no_amp_frameshift_count)
print "   MYC not amplified other mutations: "+str(MYC_no_amp_no_frameshift_count)
print "   "
print "   Fisher exact test (two-sided): "
print "      -- odd's ratio: "+str(oddsratio)
print "      -- p-value: "+str(pvalue)
print "   Fisher exact test (one-sided greater): "
print "      -- odd's ratio: "+str(oddsratio_greater)
print "      -- p-value: "+str(pvalue_greater)


    

    




 13) Test whether Deleterious mutations occur more frequently in MYC (and not CCND1) CN>4 samples
   MYC amplified deleterious mutations: 36
   MYC amplified other mutations: 97
   MYC not amplified deleterious mutations: 354
   MYC not amplified other mutations: 1008
   
   Fisher exact test (two-sided): 
      -- odd's ratio: 1.05678839769
      -- p-value: 0.836100793187
   Fisher exact test (one-sided greater): 
      -- odd's ratio: 1.05678839769
      -- p-value: 0.428760870613

In [19]:

    

# 13) test whether MYC focally amplified tumors contain deleterious TP53 mutations more frequently than non focally amplified Tumors
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   del_mut_found = False
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation":
         del_mut_found = True
         break
   if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and del_mut_found:
      MYC_amp_tp53_count += 1
   if sample.checkFocalGeneAmp("MYC") and not sample.checkFocalGeneAmp("CCND1") and not del_mut_found:
      MYC_amp_no_tp53_count += 1
   if not sample.checkFocalGeneAmp("MYC") and del_mut_found:
      MYC_no_amp_tp53_count += 1
   if not sample.checkFocalGeneAmp("MYC") and not del_mut_found:
      MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 13) test whether MYC focally amplified tumors contain deleterious TP53 mutations more frequently than non focally amplified Tumors"
print "   MYC amplified samples having deleterious TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC not amplified samples having deleterious TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC amplified samples having no deleterious TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC not amplified samples having no deleterious TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 13) test whether MYC focally amplified tumors contain deleterious TP53 mutations more frequently than non focally amplified Tumors
   MYC amplified samples having deleterious TP53 mutations: 24
   MYC not amplified samples having deleterious TP53 mutations: 333
   MYC amplified samples having no deleterious TP53 mutation: 95
   MYC not amplified samples having no deleterious TP53 mutations: 2827
   Fisher-exact test: Odds-ratio: 2.14471313419   P-value: 0.00232817585569
   Fisher-exact test (one-sided greater): Odds-ratio: 2.14471313419   P-value: 0.0017466524582

In [20]:

    

# 14) Test whether samples with copynumber > 4 in MYC (but not CCND1) contain deleterious TP53 mutations (Frameshift_indels and Nonsense) more frequently than tumors having copynumber < 4 in MYC
MYC_amp_tp53_count=0
MYC_no_amp_tp53_count=0
MYC_amp_no_tp53_count=0
MYC_no_amp_no_tp53_count=0

for sample in samples.values():
   if not sample.somatic_mutation_data:
      continue
   del_mut_found = False
   for mutation in sample.somatic_mutations:
      if mutation.gene != "TP53":
         continue
      if mutation.mut_type == "Frame_Shift_Del" or mutation.mut_type == "Frame_Shift_Ins" or mutation.mut_type == "Nonsense_Mutation":
         del_mut_found = True
         break
   copynumber_MYC = "n/a"
   log2_MYC = sample.getLog2FromGene(gene_positions['MYC'].chrom,gene_positions['MYC'].start, gene_positions['MYC'].end)
   if (log2_MYC != None):
      copynumber_MYC = 2.0 * 2.0 ** log2_MYC
   else:
      copynumber_MYC="n/a"
   copynumber_CCND1 = "n/a"
   log2_CCND1 = sample.getLog2FromGene(gene_positions['CCND1'].chrom,gene_positions['CCND1'].start, gene_positions['CCND1'].end)
   if (log2_CCND1 != None):
      copynumber_CCND1 = 2.0 * 2.0 ** log2_CCND1
   else:
      copynumber_CCND1="n/a"
   if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and del_mut_found:
      MYC_amp_tp53_count += 1
   if copynumber_MYC != "n/a" and copynumber_MYC > 4 and copynumber_CCND1 != "n/a" and copynumber_CCND1 < 4 and not del_mut_found:
      MYC_amp_no_tp53_count += 1
   if copynumber_MYC != "n/a" and copynumber_MYC < 4 and del_mut_found:
      MYC_no_amp_tp53_count += 1
   if copynumber_MYC != "n/a" and copynumber_MYC < 4 and not del_mut_found:
      MYC_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[MYC_amp_tp53_count, MYC_amp_no_tp53_count], [MYC_no_amp_tp53_count, MYC_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 14) Test whether samples with copynumber > 4 in MYC (but not CCND1) contain deleterious TP53 mutations (Frameshift_indels and Nonsense) more frequently than tumors having copynumber < 4 in MYC"
print "   MYC CN>4 and CCND1 CN<4 samples having deleterious TP53 mutations: "+str(MYC_amp_tp53_count)
print "   MYC CN<4 samples having deleterious TP53 mutations: "+str(MYC_no_amp_tp53_count)
print "   MYC CN>4 and CCND1 CN<4 having no deleterious TP53 mutation: "+str(MYC_amp_no_tp53_count)
print "   MYC CN<4 samples having no deleterious TP53 mutations: "+str(MYC_no_amp_no_tp53_count)
print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 14) Test whether samples with copynumber > 4 in MYC (but not CCND1) contain deleterious TP53 mutations (Frameshift_indels and Nonsense) more frequently than tumors having copynumber < 4 in MYC
   MYC CN>4 and CCND1 CN<4 samples having deleterious TP53 mutations: 37
   MYC CN<4 samples having deleterious TP53 mutations: 320
   MYC CN>4 and CCND1 CN<4 having no deleterious TP53 mutation: 147
   MYC CN<4 samples having no deleterious TP53 mutations: 2763
   Fisher-exact test: Odds-ratio: 2.17327806122   P-value: 0.000140653361828
   Fisher-exact test (one-sided greater): Odds-ratio: 2.17327806122   P-value: 0.000117835097918

In [21]:

    

# 15) test whether FGFR1 focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
FGFR1_amp_tp53_count=0
FGFR1_no_amp_tp53_count=0
FGFR1_amp_no_tp53_count=0
FGFR1_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  if sample.checkFocalGeneAmp("FGFR1") and not sample.checkFocalGeneAmp("ERBB2") and "TP53" in sample.genes_affected:
     FGFR1_amp_tp53_count += 1
  if sample.checkFocalGeneAmp("FGFR1") and not sample.checkFocalGeneAmp("ERBB2") and "TP53" not in sample.genes_affected:
     FGFR1_amp_no_tp53_count += 1
  if not sample.checkFocalGeneAmp("FGFR1") and "TP53" in sample.genes_affected:
     FGFR1_no_amp_tp53_count += 1
  if not sample.checkFocalGeneAmp("FGFR1") and "TP53" not in sample.genes_affected:
     FGFR1_no_amp_no_tp53_count += 1
                        
oddsratio, pvalue = scipy.stats.fisher_exact([[FGFR1_amp_tp53_count, FGFR1_amp_no_tp53_count], [FGFR1_no_amp_tp53_count, FGFR1_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[FGFR1_amp_tp53_count, FGFR1_amp_no_tp53_count], [FGFR1_no_amp_tp53_count, FGFR1_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 15) test whether FGFR1 (but not ERBB2) focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors"
print "   FGFR1 amplified samples having TP53 mutations: "+str(FGFR1_amp_tp53_count)
print "   FGFR1 not amplified samples having TP53 mutations: "+str(FGFR1_no_amp_tp53_count)
print "   FGFR1 amplified samples having no TP53 mutation: "+str(FGFR1_amp_no_tp53_count)
print "   FGFR1 not amplified samples having no TP53 mutations: "+str(FGFR1_no_amp_no_tp53_count)
print "   Fisher-exact test: Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)


    

    




 15) test whether FGFR1 (but not ERBB2) focally amplified tumors contain TP53 mutations more frequently than non focally amplified Tumors
   FGFR1 amplified samples having TP53 mutations: 118
   FGFR1 not amplified samples having TP53 mutations: 1171
   FGFR1 amplified samples having no TP53 mutation: 113
   FGFR1 not amplified samples having no TP53 mutations: 1875
   Fisher-exact test: Odds-ratio: 1.67204492038   P-value: 0.000201890837143
   Fisher-exact test (one-sided greater): Odds-ratio: 1.67204492038   P-value: 0.000115492642829

In [22]:

    

# 16) test whether tumors with copynumber > 4 in FGFR1 but copynumber < 4 in ERBB2 contain TP53 mutations more frequently than tumors with FGFR1 CN<4
FGFR1_amp_tp53_count=0
FGFR1_no_amp_tp53_count=0
FGFR1_amp_no_tp53_count=0
FGFR1_no_amp_no_tp53_count=0

for sample in samples.values():
  if not sample.somatic_mutation_data:
     continue
  copynumber_FGFR1 = "n/a"
  log2_FGFR1 = sample.getLog2FromGene(gene_positions['FGFR1'].chrom,gene_positions['FGFR1'].start, gene_positions['FGFR1'].end)
  if (log2_FGFR1 != None):
     copynumber_FGFR1 = 2.0 * 2.0 ** log2_FGFR1
  else:
     copynumber_FGFR1="n/a"
  copynumber_ERBB2 = "n/a"
  log2_ERBB2 = sample.getLog2FromGene(gene_positions['ERBB2'].chrom,gene_positions['ERBB2'].start, gene_positions['ERBB2'].end)
  if (log2_ERBB2 != None):
     copynumber_ERBB2 = 2.0 * 2.0 ** log2_ERBB2
  else:
     copynumber_ERBB2="n/a"
  if copynumber_FGFR1 != "n/a" and copynumber_FGFR1 > 4 and copynumber_ERBB2 != "n/a" and copynumber_ERBB2 < 4 and "TP53" in sample.genes_affected:
     FGFR1_amp_tp53_count += 1
  if copynumber_FGFR1 != "n/a" and copynumber_FGFR1 < 4 and "TP53" in sample.genes_affected:
     FGFR1_no_amp_tp53_count += 1
  if copynumber_FGFR1 != "n/a" and copynumber_FGFR1 > 4 and copynumber_ERBB2 != "n/a" and copynumber_ERBB2 < 4 and "TP53" not in sample.genes_affected:
     FGFR1_amp_no_tp53_count += 1
  if copynumber_FGFR1 != "n/a" and copynumber_FGFR1 < 4 and "TP53" not in sample.genes_affected:
     FGFR1_no_amp_no_tp53_count += 1
oddsratio, pvalue = scipy.stats.fisher_exact([[FGFR1_amp_tp53_count, FGFR1_amp_no_tp53_count], [FGFR1_no_amp_tp53_count, FGFR1_no_amp_no_tp53_count]])
oddsratio_greater, pvalue_greater = scipy.stats.fisher_exact([[FGFR1_amp_tp53_count, FGFR1_amp_no_tp53_count], [FGFR1_no_amp_tp53_count, FGFR1_no_amp_no_tp53_count]], alternative='greater')

print ""
print " 16) test whether tumors with copynumber > 4 in FGFR1 but copynumber < 4 in ERBB2 contain TP53 mutations more frequently than tumors with FGFR1 CN<4 Tumors"
print "   FGFR1 CN>4, ERBB2 CN<4 samples having TP53 mutations: "+str(FGFR1_amp_tp53_count)
print "   FGFR1 CN<4 samples having TP53 mutations: "+str(FGFR1_no_amp_tp53_count)
print "   FGFR1 CN>4, ERBB2 CN<4  samples having no TP53 mutation: "+str(FGFR1_amp_no_tp53_count)
print "   FGFR1 CN<4 samples having no TP53 mutations: "+str(FGFR1_no_amp_no_tp53_count)
print "   Fisher-exact test (two-sided): Odds-ratio: "+str(oddsratio)+"   P-value: "+str(pvalue)
print "   Fisher-exact test (one-sided greater): Odds-ratio: "+str(oddsratio_greater)+"   P-value: "+str(pvalue_greater)