Mapeamento genético em famílias de meio-irmãos por simulação computacional

• Autor(a) principal: God, Pedro Ivo Vieira Good
• Data de Publicação: 2008
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/1285
• topic: Famílias de meio-irmãos; QTL; Simulação; Half-sib families; QTL; Simulation; CNPQ::CIENCIAS BIOLOGICAS::GENETICA::GENETICA QUANTITATIVA

This study aimed at investigating the genetic mapping in half- sib families (HSF) through computer simulation to elucidate its limitations and advantages in genomic analysis. The following scenarios of simulation were used: (1) Study the effects of population size in HSF for N = 50, 100, 200, 300, 500 and 1000; (2) Study the effects of saturation degrees (5 cM, 10 cM and 15 cM) of the genome with molecular markers; (3) Study the effects of levels of information by polymorphism information content (PIC), with s = 2, 3, 4, 5 and 6 alleles equifrequents in the mapping populations. All the possible combinations among these various scenarios were simulated in 100 repetitions each. This generated a total of 3000 populations for analysis. We used the following criteria for determining the accuracy of genetic mapping: (i) number of linkage groups recovered, (ii) sizes of linkage groups, (iii) average distance between adjacent markers in linkage groups, (iv) variance of the distances between adjacent markers in the linkage groups, (v) stress, and (vi) Spearman's correlation. The results show that besides population size and degree of saturation, PIC has a critical role in the mapping of HSF. Populations with N = 50 and 100 are not recommended for the mapping in HSF, for all levels of PIC. When the level of PIC in the population is minimal (s = 2), population sizes of N <300 are not sufficient to obtain reliable maps, especially for those with a low degree of saturation, in which case there occurs an increase in inversions of molecular markers in linkage map. Maps less saturated showed lower genome recovery, lower estimates of correlation of Spearman and higher variance of distances between adjacent markers, especially for low level of PIC. As a general conclusion, we do not recommended the use of populations with N = 50 and 100 even with a high level of PIC. For N = 200 one can get maps with some fidelity provided that the number of segregating alleles in the base population is equal to or greater than 4, similarly to that which occurs in Full-Sib Families from matings between fully informative parents.