The most important product of the sequencing of a genome is a complete, acc
urate catalogue of genes and their products, primarily messenger RNA transc
ripts and their cognate proteins. Such a catalogue cannot be constructed by
computational annotation alone; it requires experimental validation on a g
enome scale. Using 'exon' and 'tiling' arrays fabricated by ink-jet oligonu
cleotide synthesis, we devised an experimental approach to validate and ref
ine computational gene predictions and define full-length transcripts on th
e basis of co-regulated expression of their exons. These methods can provid
e more accurate gene numbers and allow the detection of mRNA splice variant
s and identification of the tissue- and disease-specific conditions under w
hich genes are expressed. We apply our technique to chromosome 22q under 69
experimental condition pairs, and to the entire human genome under two exp
erimental conditions. We discuss implications for more comprehensive, consi
stent and reliable genome annotation, more efficient, full-length complemen
tary DNA cloning strategies and application to complex diseases.