Prediction of BRCAness/homologous recombination deficiency of breast tumors on digitalized slides

A computer-implemented method for visualizing clusters of sub-images or tiles of at least one biological class, of at least one biological image, comprising the following steps:

• a. dividing the image into sub-images or tiles,
• b. selecting at least some of the tiles from the set of tiles by removing the background tiles,
• c. encoding each tile or each selected tile, via a pre-trained convolutional neural network, so as to obtain a representation vector or tensor for each tile;
• d. assigning a score, also called attention score, to each tile,
• e. selecting tiles based on the attention score by selecting the tiles with the highest attention scores,
• f. assigning a score, also called decision score, to each tile by predicting the output class from each individual tile,
• g. further selecting tiles, as to keep only tiles that have both a high attention and a high decision score,
• h. projecting the tile representation of said tiles or said selected tiles to a a 2-dimensional or 3-dimensional space by using the U-MAP (Uniform Manifold Approximation and Projection) or T-SNE (T-distributed Stochastic Neighbor Embedding) algorithm,
• wherein the biological class is the genomic signature or profile of the cancer, or a molecular class of cancer, selected from triple negative breast cancer or luminal breast cancer, or the class is selected from the cancer's Grade, from the gBRCA1/2 (germline) status,
• selected from sporadic or germinal cancer, or from the homologous recombination status of a cancer,
• wherein the biological class is the genomic tumor (or cancer) profile, a Homologous Recombination Deficient (HRD) profile, defined by the presence of a germline BRCA1/2 (gBRCA1/2) mutation or assessed by Large-scale State Transitions (LST) genomic signature.