+// Wigginton 2005, PMID: 15789306
+// written by Jan Wigginton
+double calc_hwe(int obs_hom1, int obs_hom2, int obs_hets)
+{
+ if (obs_hom1 + obs_hom2 + obs_hets == 0 ) return 1;
+
+ assert(obs_hom1 >= 0 && obs_hom2 >= 0 && obs_hets >= 0);
+
+ int obs_homc = obs_hom1 < obs_hom2 ? obs_hom2 : obs_hom1;
+ int obs_homr = obs_hom1 < obs_hom2 ? obs_hom1 : obs_hom2;
+
+ int rare_copies = 2 * obs_homr + obs_hets;
+ int genotypes = obs_hets + obs_homc + obs_homr;
+
+ double *het_probs = (double*) calloc(rare_copies+1, sizeof(double));
+
+ /* start at midpoint */
+ int mid = rare_copies * (2 * genotypes - rare_copies) / (2 * genotypes);
+
+ /* check to ensure that midpoint and rare alleles have same parity */
+ if ((rare_copies & 1) ^ (mid & 1)) mid++;
+
+ int curr_hets = mid;
+ int curr_homr = (rare_copies - mid) / 2;
+ int curr_homc = genotypes - curr_hets - curr_homr;
+
+ het_probs[mid] = 1.0;
+ double sum = het_probs[mid];
+ for (curr_hets = mid; curr_hets > 1; curr_hets -= 2)
+ {
+ het_probs[curr_hets - 2] = het_probs[curr_hets] * curr_hets * (curr_hets - 1.0) / (4.0 * (curr_homr + 1.0) * (curr_homc + 1.0));
+ sum += het_probs[curr_hets - 2];
+
+ /* 2 fewer heterozygotes for next iteration -> add one rare, one common homozygote */
+ curr_homr++;
+ curr_homc++;
+ }
+
+ curr_hets = mid;
+ curr_homr = (rare_copies - mid) / 2;
+ curr_homc = genotypes - curr_hets - curr_homr;
+ for (curr_hets = mid; curr_hets <= rare_copies - 2; curr_hets += 2)
+ {
+ het_probs[curr_hets + 2] = het_probs[curr_hets] * 4.0 * curr_homr * curr_homc /((curr_hets + 2.0) * (curr_hets + 1.0));
+ sum += het_probs[curr_hets + 2];
+
+ /* add 2 heterozygotes for next iteration -> subtract one rare, one common homozygote */
+ curr_homr--;
+ curr_homc--;
+ }
+ int i;
+ for (i = 0; i <= rare_copies; i++) het_probs[i] /= sum;
+
+ /* p-value calculation for p_hwe */
+ double p_hwe = 0.0;
+ for (i = 0; i <= rare_copies; i++)
+ {
+ if (het_probs[i] > het_probs[obs_hets])
+ continue;
+ p_hwe += het_probs[i];
+ }
+
+ p_hwe = p_hwe > 1.0 ? 1.0 : p_hwe;
+ free(het_probs);
+ return p_hwe;
+
+}
+
+
+static void _bcf1_set_ref(bcf1_t *b, int idp)
+{
+ kstring_t s;
+ int old_n_gi = b->n_gi;
+ s.m = b->m_str; s.l = b->l_str - 1; s.s = b->str;
+ kputs(":GT", &s); kputc('\0', &s);
+ b->m_str = s.m; b->l_str = s.l; b->str = s.s;
+ bcf_sync(b);
+
+ // Call GTs
+ int isample, an = 0;
+ for (isample = 0; isample < b->n_smpl; isample++)
+ {
+ if ( idp>=0 && ((uint16_t*)b->gi[idp].data)[isample]==0 )
+ ((uint8_t*)b->gi[old_n_gi].data)[isample] = 1<<7;
+ else
+ {
+ ((uint8_t*)b->gi[old_n_gi].data)[isample] = 0;
+ an += b->ploidy ? b->ploidy[isample] : 2;
+ }
+ }
+ bcf_fit_alt(b,1);
+ b->qual = 999;
+
+ // Prepare BCF for output: ref, alt, filter, info, format
+ memset(&s, 0, sizeof(kstring_t)); kputc('\0', &s);
+ kputs(b->ref, &s); kputc('\0', &s);
+ kputs(b->alt, &s); kputc('\0', &s); kputc('\0', &s);
+ {
+ ksprintf(&s, "AN=%d;", an);
+ kputs(b->info, &s);
+ anno16_t a;
+ int has_I16 = test16(b, &a) >= 0? 1 : 0;
+ if (has_I16 )
+ {
+ if ( a.is_tested) ksprintf(&s, ";PV4=%.2g,%.2g,%.2g,%.2g", a.p[0], a.p[1], a.p[2], a.p[3]);
+ ksprintf(&s, ";DP4=%d,%d,%d,%d;MQ=%d", a.d[0], a.d[1], a.d[2], a.d[3], a.mq);
+ }
+ kputc('\0', &s);
+ rm_info(&s, "I16=");
+ rm_info(&s, "QS=");
+ }
+ kputs(b->fmt, &s); kputc('\0', &s);
+ free(b->str);
+ b->m_str = s.m; b->l_str = s.l; b->str = s.s;
+ bcf_sync(b);
+}
+
+int call_multiallelic_gt(bcf1_t *b, bcf_p1aux_t *ma, double threshold, int var_only)
projects / samtools.git / commitdiff