Table 3. Comparison of read mappings to reference genome and paternal and maternal haplotypes of GM12878.

Haplotype	No. of mapped reads	Reference	Paternal	Maternal
Chip-Seq reads for Pol II were independently mapped to each haplotype (chromosomes 1–22 and X) and the best unambiguous mapping (no more than two mismatches) was selected for each read. More reads are mapped to either haplotypes of GM12878 than to the reference genome. The major difference in numbers for paternal/maternal and reference haplotypes is due to reads that map to one haplotype but not to other. Namely, only about 0.1–0.2% of reads that map to the reference cannot be mapped to paternal/maternal haplotype, while a significantly higher fraction ∼0.5% of reads map to paternal/maternal genome and cannot be mapped to the reference. Interestingly, for paternal and maternal haplotypes unmapped reads and reads with different mappings contribute roughly equally to the discrepancy in overall mapping. See Supplementary Table 1 for the results for CTCF.
		Equivalently mapped reads in
Reference	69 086 591		68 942 501 (99.79%)	69 034 357 (99.92%)
Paternal	(+0.3%) 69 296 783	68 942 501 (99.49%)		69 099 705 (99.72%)
Maternal	(+0.4%) 69 394 995	69 034 357 (99.48%)	69 099 705 (99.58%)
		Differently mapped reads in
Reference	69 086 591		18 248 (0.03%)	18 291 (0.03%)
Paternal	(+0.3%) 69 296 783	18 248 (0.03%)		113 796 (0.16%)
Maternal	(+0.4%) 69 394 995	18 291 (0.03%)	113 796 (0.16%)
		Unmapped reads in
Reference	69 086 591		125 842 (0.18%)	33 943 (0.05%)
Paternal	(+0.3%) 69 296 783	336 034 (0.48%)		83 282 (0.12%)
Maternal	(+0.4%) 69 394 995	342 347 (0.49%)	181 494 (0.26%)