TAILIEUCHUNG - Báo cáo sinh học: "Bias in multiple genetic correlation from half-sib designs"

Tuyển tập các báo cáo nghiên cứu về sinh học được đăng trên tạp chí sinh học Journal of Biology đề tài: Bias in multiple genetic correlation from half-sib designs | 335 Genet Sei Evol 1995 27 335-345 Elsevier INRA Original article Bias in multiple genetic correlation from half-sib designs PM Visscher Roslin Institute Edinburgh Roslin Midlothian EH25 9PS UK Received 6 June 1994 accepted 15 March 1995 Summary - Mean squares and mean crossproducts between and within sires were simulated to investigate the bias in genetic R2 defined as the square of the multiple correlation between a single trait and q 1 other traits calculated from an estimate of the genetic covariance matrix from balanced half-sib designs. Approximate prediction equations for this bias were derived when the population correlation was zero. In that case the bias is approximately inversely proportional to the degrees of freedom for estimating sire components and the reliabilities of the implicit progeny test and proportional to g-1 . Using a genetic multiple regression based on a large number of traits and or a small number of sires could lead to loss in response to selection relative to using a regression based on the true population parameters. genetic regression bias multiple correlation REML half-sib design Resume - Biais de la correlation génétique multiple R2 dans un plan experimental avec demi-freres. Des Carres et des co-produits moyens entre pères et intra-père ont été simulés pour étudier le biais du R2 génétique defini comme le carré de la correlation multiple entre un caractère et q 1 autres caractères calculée à partir d une estimée de la matrice des covariances génétiques dans des schemas expérimentaux équilibrés comprenant des demi-freres. Des equations approximatives de prediction de ce biais ont été établies dans le cas d une correlation nulle dans la population. Dans le cas le biais est à peu près inversement proportionnel aux degrés de liberté d estimation des composantes paternelles à la precision de répreuve de descendance implicite et proportionnel à q - 1 . Si on utilise une regression génétique multiple basée sur un grand nombre de .

• Báo cáo khoa học
• Báo cáo sinh học hay
• cách trình bày báo cáo
• báo cáo sinh học
• công trình nghiên cứu sinh học
• tài liệu về sinh học