The human Chromosome 22 placental transcriptome. Dark blue strips indicate regions that are represented on the Chromosome 22 DNA microarray. Red strips indicate the positions of Sanger Centre release 2.3 annotated genes. The magnitude of the density plot represents the number of positive hybridizing fragments divided by the total number of fragments in a 100-kb window. (A) A large amount of transcriptional activity in a previously unannotated region of Chromosome 22. (B) A peak in transcriptional activity corresponding to known gene annotations. The window was moved fragment by fragment to give a continuous density plot. Positions with spikes in the density plot and low frequency of red strips indicate regions of novel transcriptional activity. Coordinates are given from centromere to telomere (starting at band 22q11.1) because the p arm has not been sequenced. The NCBI assembly lists sequence coordinates from the 5′ end of the p telomere to the 3′ end of the q telomere. Adding a 5′ offset of exactly 13 Mb to approximate the size of the p arm establishes a common reference frame for the NCBI/UCSC Golden Path assembly.

Northern blot analysis of 118 fragments that were expressed in previously unannotated regions of Chromosome 22. Thirty (25.4%) showed discrete bands. Ten are shown above and labeled with the corresponding chromosomal location of the probe used in the Northern hybridization. Bar indicates two probes separated by ∼30 kb in genomic space that hybridize to the same 6-kb transcript.

Differential hybridization mapping within positive PCR fragment sequences. (A) Hybridization to multiple 60-nt oligonucleotides positioned opposite an intron sequence annotated on the antisense strand. (B) Hybridization to oligonucleotides representing a predicted exon within an annotated intron on the sense strand. (C) Control spots showing differential hybridization to a known exon (1) located on the strand opposite an annotated intron and (2) whose expression was previously verified. NCBI/UCSC sequence coordinates are offset by 13 Mb to approximate the size of the unsequenced p arm.

Mouse protein homology within translated PCR fragment sequences. (A) Homology match between mouse sequence and a positive microarray fragment whose sequence coordinates lie outside annotated genes. (B,C) Examples of mouse protein matches to human genomic sequences that are opposite annotated introns. In both cases the homology match is antisense to the intron. NCBI/UCSC sequence coordinates are offset by 13 Mb to approximate the size of the unsequenced p arm.

False positives. Determination of intensity cutoff in determining positive hybridizing fragments. There is a clear leveling of consistency after 200 intensity units. The plot demonstrates that fragments with an intensity >200 were present with that intensity or higher in 5 out 6 replicate experiments. Fragments that hybridized in 4 or less of the 6 replicate experiments and with an intensity ≥200 were summed to give a false-positive rate of 5%.