| Specific language impairment is characterized as a failure to develop language normally, in the absence of explanatory factors such as hearing deficits, low IQ, extreme neglect, epilepsy or gross neurological impairment. Subjects with SLI show poor performance on phonologically based tasks and proper use of grammatical morphemes. Several theories of the etiology of SLI have been put forward in the literature but no consensus for the cause of SLI is agreed upon. Previous familial aggregation studies have shown that SLI clusters in families and twin studies have further indicated that much of the correlation is due to genetics. However, no genetic factors have been isolated to date.; The work presented in this dissertation includes two full genome scans for SLI susceptibility loci. A sample of multiplex nuclear and extended pedigrees from the US (n = 22) were ascertained on the basis of 2 pedigree members classified as SLI. These pedigrees were characterized by several neuropsychological measures of language, reading and IQ and assessed for comorbidity of language and reading deficits and sex effects. A second sample of multiplex nuclear and extended pedigrees from Canada (n = 5) were ascertained on the basis of at least 2 pedigree members classified as SLI and phenotypically characterized using a similar battery.; Genome-wide linkage analysis was conducted using standard LOD scores and a Bayesian linkage method, the posterior probability of linkage (PPL). Three affection statuses were employed based on language, reading or clinical impairments. The strongest evidence for linkage was seen on 13q21--22 with a PPL of 53% in the Canadian sample and a PPL of 92% when combined with the US sample. The linkage region on 13q21 has also been implicated in samples ascertained for autism, primarily attributable to families that show severe language delay (onset of phrase speech >36 months).; Several other loci were suggested in the course of analysis, though no linkage was observed in regions implicated by another research group. Their ascertainment, pedigree structures and analysis methods differ substantially from the present work. The potential of different analysis methods in the context of SLI is discussed and alternative approaches to analysis of the current data set are examined. This dissertation represents a critical step in localizing SLI susceptibility genes and may provide more detailed information about the neurobiology of language development in the future. |
