Abstract: The disclosure notably relates to a computer-implemented method for simulating evolution of a tumor associated to an oncogene. The method includes providing a plurality of pieces of data, each corresponding to a given cell of the tumor, and includes a degree of activation of the oncogene in the given cell. The method further includes providing a model configured to take an input piece of data and to output information on proliferation of the respective given cell corresponding to the input piece of data. The information on proliferation depends on the degree of activation of the oncogene. The method further includes running the model on one or more pieces of data of the plurality of pieces of data and updating the plurality of pieces of data based on the result of the running. Such a method improves the simulation of the evolution of a tumor.

Abstract: A computer-implemented method for machine-learning a neural network for variant calling with respect to a reference genome. The neural network takes as input one or more sets of data pieces each specifying a respective read aligned relative to a genomic position of the reference genome. The neural network outputs information with respect to presence of a variant at the genomic position. The neural network includes, for each set of data pieces, a respective function configured to take as input and process the set of data pieces. The respective function is symmetric. The machine-learning method improves variant calling with respect to a reference genome.

Abstract: A computer-implemented for parameterizing a statistical function. The statistical function is configured to perform prognosis on cancer patients. The parameterizing method includes obtaining a dataset relative to a plurality of patients each having a cancer disease. The dataset includes, for each patient, input data of the statistical function. The input data include visual data from at least one histological image of a tumor slide of the patient. The input data also include genomic data from sequencing tumor tissue of the patient. The input data further include clinical data of the patient. The dataset also includes clinical endpoints of the patient with respect to evolution of the cancer disease. The parameterizing method further includes parameterizing the statistical function based on the dataset. This forms an improved solution to perform prognosis on patients having a cancer disease.

Abstract: A computer-implemented method for machine-learning a function configured, based on input covariates representing medical characteristics of a patient with respect to a multi-state model of an illness having states and transitions between the states, to output a distribution of transition-specific probabilities for each interval of a set of intervals, the set of intervals forming a subdivision of a follow-up period. The machine-learning method including obtaining a dataset of covariates and time-to-event data of a set of patients, and training the function based on the dataset. This forms an improved solution for determining accurate patient data with respect to a multi-state model of an illness.

Abstract: The disclosure notably relates to a computer-implemented method for simulating evolution of a tumor associated to an oncogene. The method includes providing a plurality of pieces of data, each corresponding to a given cell of the tumor, and includes a degree of activation of the oncogene in the given cell. The method further includes providing a model configured to take an input piece of data and to output information on proliferation of the respective given cell corresponding to the input piece of data. The information on proliferation depends on the degree of activation of the oncogene. The method further includes running the model on one or more pieces of data of the plurality of pieces of data and updating the plurality of pieces of data based on the result of the running. Such a method improves the simulation of the evolution of a tumor.