Applications of fluorescence in situ hybridization (FISH) for transloc
ation studies and biological dosimetry would benefit substantially fro
m reliable and efficient automatic detection of metaphase chromosomes
labeled with fluorescent dyes. We replicated and evaluated a fluoresce
nce metaphase finder previously developed at the Medical Research Coun
cil (MRC), Human Genetics Unit (Scotland) and at Lawrence Berkeley Lab
oratory (LBL; California), The MRC/LBL system seemed to detect nearly
all of the metaphases on the test slides, but it presented an unaccept
able number of false positives (about five false positives per one tru
e positive). Furthermore, we determined that the system actually overc
alled true detections by counting certain metaphase spreads twice (dup
licates), Through modifications of the MRC/LBL system, we developed th
e Lawrence Livermore National Laboratory (LLNL) system, which minimize
s the detection of duplicates, incorporates new detection features, us
es a binary decision tree (BDT) for classification, and provides funct
ionalities to improve scanning accuracy and improve the post-detection
review. To test the new system, DAPI-stained preparations of metaphas
e chromosomes from blood lymphocytes of four unrelated donors were pla
ced on slides in drops ranging from 7 mm to 20 mm in diameter. Drops c
ontained between 5 and 200 scorable metaphases each, The LLNL system a
chieved similar to 90% detection of non-duplicated metaphases as verif
ied by an expert cytogeneticist, with typically less than one false po
sitive per every one true positive detected. (C) 1998 Wiley-Liss, Inc.
dagger.