Retinitis pigmentosa (RP) defines a group of hereditary retinal dystrophies characterised by a progressive deterioration of night vision and reduction of the visual field due to photoreceptor degeneration. RP shows both clinical and genetic heterogeneity and has X- linked, autosomal recessive and autosomal dominant forms of inheritance. In addition, a 'digenic' form of inheritance had also been reported. To date, positional cloning and candidate gene approaches have identified more than 20 genes and gene loci responsible for RP, and the number is increasing. In the present study, three Scottish RP families were included in a linkage study, to determine if the disease locus in each of these families mapped to any of the known RP loci on chromosomes 7 (7p and 7q), 8 and 19. Significant positive LOD scores were obtained for family G-adRP, with markers mapping within the RPIO locus (7q31-q32). The highest LOD score obtained was 3.913 (at 0% recombination) with marker D7S514. Results for the remaining two families were generally inconclusive. However, LOD scores obtained indicated that linkage to 7p and 7q was significantly excluded for family F-adRP. No conclusive exclusion of linkage was obtained for family C-RP. Two candidate genes exist within the RPIO candidate region, namely the blue cone pigment gene (BCP) and the metabotropic glumate receptor 8 (GRM8) gene. Mutation screening was performed using SSCP analysis to evaluate the involvement of these genes in the pathogenesis of RP in family G-adRP. An A->C polymorphism was observed in the final base of codon 122 of the BCP gene (exon 2). The amino acid residue (glycine) is not altered, however, and is thus unlikely to be involved in the disease process. Furthermore, this polymorphism was noted in both affected and unaffected individuals of family G-adRP, as well as in control individuals. For the GRM8 gene, only eight out of ten exons were available for analysis at the time of study. SSCP mobility shifts were detected in 4 PCR fragments (exons IV, V and X, and exon VIIIb). Sequencing analysis of exon VIIIb revealed an A-C polymorphism in the first base of codon 561. The involvement of this polymorphism in the disease process is unlikely as the change does not alter the encoded amino acid (arginine) and it was detected in both affected and unaffected members of G- adRP. In exon X, a heterozygous C>T sequence transition was noted 29bp downstream of the stop codon. However, the shift was again found in both affected and unaffected individuals of family G-adRP, therefore unlikely to be pathogenic. Sequence analysis of the remaining two exons (exons IV and V) did not reveal any deviation from the published control sequence. The direct cDNA selection, technique was employed in the attempt to isolate transcripts from the candidate region. YACs spanning the region of interest were obtained and their inserts verified by FISH and PCR-based STS content screening. Selection was performed on human retinal cDNA library preparations, and the selected products were subcloned to create a selected-cDNA library for further analysis. Sequencing analysis of a subset (27clones) of the selected products indicated that some fragments were indeed derived from within the 7q3 l-q32 region, as shown by BLAST and FASTA sequence homology searches, although most were repetitive in origin. Hits to known genes or other transcribed sequences (ESTs or cloned cDNA) were also obtained, although several of these genes have been shown to map elsewhere in the genome. At present, family G-adRP is the fourth autosomal dominant RP family whose disease locus was mapped within the RPIO region (7q31-q32). All known candidate genes identified to date have been excluded, which necessitates positional cloning strategies to be employed. Using the direct cDNA selection technique, a library of cDNA fragments putatively isolated from a portion of the RPIO candidate region has been constructed in this study. However, time was a limiting factor, and as a result, only a minor subset (less than 15%) of the selected-cDNA library was analysed. Further analysis and characterisation of the selected cDNAs is warranted to identify clones that represent (novel) transcripts mapping within the 7q31-q32 candidate region, which will undoubtedly expedite the identification of the RPIO gene.