In this project, we shall generate and evaluate three chromosome 5- specific cDNA libraries by direct selection: a technique that is capable of isolating even low abundance cDNAs. The selected cDNAs will be used en masse to screen an arrayed set of chromosome 5-specific genomic clones, thus identifying at least some of the transcribed genomic regions. The cDNAs will also be built into an arrayed high- density set of 50,000 clones that will be screened with genomic YAC and cosmid contigs as they are developed, thus adding expression data to the expanding physical map and directly identifying cDNAs that are expressed by genomic contigs. 500 chromosome 5-specific selected cDNAs will be converted to Expressed Sequence Tags (ESTs), placed on the physical map using radiation hybrid mapping, and used to identify their homologous genomic sequences. Using these methods, we will be able to annotate the cosmid, YAC, and radiation hybrid physical maps with a large number of genes and be able to anchor the various maps together. Applications of multiplex selection and cDNA normalization strategies should also prove useful in the rapid derivation of tissue specific transcription maps across large regions of the human genome and in assessing the distribution of coding sequences throughout chromosome 5.