Comparative genomic hybridization (CGH) allows entire genomes to be sc
anned for whole and segmental aneuploidy and thus may be an appropriat
e tool for the detection of clinically important abnormalities during
fetal and neonatal screening, Criteria to distinguish between signific
ant aberrations and experimental artifacts are essential for these app
lications, This report describes the use of a t-statistic to detect ch
anges in CGH profiles that differ significantly from variations that o
ccur in CGH profiles of normal samples. Eleven cell Lines derived from
fetal or neonatal patients were analyzed in this study, Aneuploidies
in these Lines included trisomies for chromosomes 13, 16, 18, and 21 a
nd monosomy for distal 5p and tetrasomy 18p. Aneuploidy was detected i
n all samples by using the t-statistic, although the extent of the ane
uploid region was not correctly estimated in some cases. A detailed de
scription of the t-statistic used for making these CGH comparisons is
described in a companion paper (Moore et al., Cytometry 28:183-130, 19
97). (C) 1997 Wiley-Liss, Inc.