Asymmetric trichotomous partitioning overcomes dataset limitations in building machine learning models for predicting siRNA efficacy.

There was an overall trend of higher performance for models built with the most stringent threshold pairs (small h 1 /large h 2 ; Figure 5 , top left curves), while models built with all other threshold combinations performed poorly ( Figure 5 , bottom left, top right, and center curves).

also in 1,675 other papers