. Author manuscript; available in PMC: 2012 Dec 6.

Published in final edited form as: Sci Transl Med. 2012 Jun 6;4(137):137ra76. doi: 10.1126/scitranslmed.3004323

Table 2.

Number of sites and accuracy of fetal genotype inference from maternal plasma sequencing (percentage of transmitted alleles correct out of all predicted) by parental genotype and phasing status. Sites later determined by trio sequencing (including the offspring) to have poor genotype quality scores or genotypes that violated Mendelian inheritance were discarded the purpose of evaluating accuracy (14,000 maternal-only, 32,233 paternal-only, and 480 shared heterozygous sites, or 1.5% of all sites).

Individual	Site	Other parental genotype	Number of Sites	Accuracy
Mother (I1-M)	Heterozygous, phased	Homozygous	1,064,255	99.3%
	Heterozygous, phased	Heterozygous	576,242	98.7%^†

	Heterozygous, not phased	All	121,425	N.D.

Father (I1-P)	Heterozygous	Homozygous	1,134,192	96.8%
Father (I1-P)	Heterozygous	Heterozygous	631,721	N.D.

^†

Among biparentally heterozygous sites, accuracy was assessed only where the offspring was homozygous (48.8%, n=631,721), allowing the “true” transmitted alleles to be unambiguously inferred from trio genotypes.