The invention discloses an organ-like
drug resistance prediction method and device and a storage medium, and relates to the technical field of
bioinformatics, and the method comprises the steps: S1, constructing an initial prediction model; s2, acquiring a training
data set; s3, training an optimization prediction model; s4, acquiring related data of a to-be-predicted
drug; s5, obtaining a prediction result of the
drug resistance of the to-be-predicted drug; according to the method, the
drug resistance of the
organoid to the drug can be predicted more accurately. By fusing multi-dimensional features of
gene mutation, target
mutation and drug functional groups, the understanding and learning ability of the model to the
drug resistance mechanism is enhanced, and the reliability of
drug resistance prediction is improved. The method not only can accurately capture key information of influence of
gene and target
mutation on drug resistance, but also can highlight unique
mutant genes of the
organoid on drug resistance by comparing feature differences of the
organoid and a common
cell line, so that an efficient and accurate calculation framework is provided for drug resistance detection.