01906naa a2200241 a 450000100080000000500110000800800410001902400440006010000220010424501000012626000090022652012040023565000110143965000200145065300220147070000190149270000160151170000230152770000190155070000210156970000240159077300500161421519592023-02-27       2022    bl uuuu     u00u1 u    #d7 ahttps://doi.org/10.1002/agj2.211472DOI1 aSANTOS, I. G. DOS  aA novel approach to determine tropical persistence on alfalfa germplasm.h[electronic resource]  c2022  aPersistence plays a key role in alfalfa (Medicago sativa L. ssp. sativa) cultivation in tropical areas, but it is still a restriction for breeding programs. The objectives of this study were to identify persistent alfalfa accessions evaluated under tropical conditions, and to propose a method for selecting persistent accessions based on random regression (RR) models using artificial neural networks (ANN). Dry matter yield (DMY) of 77 alfalfa accessions from 24 cuts was measured to evaluate persistence using different RR models. A persistence method was proposed based on the trajectory curves of the accessions. The fitted curves showed a great amplitude regarding DMY over time, which suggest high persistence variability. The three-step method for accessing persistence presented in this study included RR modeling to obtain trends of persistence, k-means to define different persistence clusters, and ANN to perform persistence classification in an automated way. When new accessions are evaluated by an alfalfa breeding program, they will be classified according to their genetic values scores using the same ANN previously fitted. The ANN will greatly enhance the decision-making process.  aAlfafa  aMedicago Sativa  aAlfalfa germplasm1 aPEIXOTO, M. A.1 aCRUZ, C. D.1 aFERREIRA, R. de P.1 aNASCIMENTO, M.1 aROSADO, R. D. S.1 aSANT ANNA, I. DE C.  tAgronomy Journalgv. 114, p. 3225-3233, 2022.