Abstract: A method may include determining, within an image of a biological sample, a first plurality of tiles having a first tile size and a second plurality of tiles having a second tile size. A feature extraction model may be applied to extract features from the different size tiles. Concatenated feature sets, each of which including a first feature of a first tile from the first plurality of tiles, a second feature of a second tile from the second plurality of tiles, and a third feature of a third tile from the second plurality of tiles, may be formed. Molecular features present in the biological sample may be determined based on an attention-weighted position embedding of the concatenated feature sets and a joint representation of features across clusters of spatially proximate tiles in the image. Related systems and computer program products are also provided.

Abstract: The present invention provides diagnostic methods, therapeutic methods, and compositions for the treatment of cancer (e.g., a bladder cancer (e.g., UC, e.g., mUC), a kidney cancer, a lung cancer, a liver cancer, an ovarian cancer, a pancreatic cancer, a colorectal cancer, or a breast cancer). The invention is based, at least in part, on the discovery that expression levels of one or more biomarkers described herein in a sample from an individual having cancer can be used in methods of identifying an individual having a cancer who may benefit with an anti-cancer therapy that includes an immunotherapy (e.g., a PD-L1 axis binding antagonist such as an anti-PD-L1 antibody (e.g., atezolizumab)) and a suppressive stromal antagonist (e.g.

Abstract: The present invention provides diagnostic methods, therapeutic methods, and compositions for the treatment of cancer (e.g., a bladder cancer (e.g., UC, e.g., mUC), a kidney cancer, a lung cancer, a liver cancer, an ovarian cancer, a pancreatic cancer, a colorectal cancer, or a breast cancer). The invention is based, at least in part, on the discovery that expression levels of one or more biomarkers described herein in a sample from an individual having cancer can be used in methods of identifying an individual having a cancer who may benefit with an anti-cancer therapy that includes an immunotherapy (e.g., a PD-L1 axis binding antagonist such as an anti-PD-L1 antibody (e.g., atezolizumab)) and a suppressive stromal antagonist (e.g.

Abstract: Embodiments disclosed herein generally relate to classifying a tumor, based on gene expression data, as being neurally related or non-neurally related. The tumor may be classified using a machine-learning model, which may have been trained to differentiate gene-expression data associated with neuronal or neuroendocrine tumors from gene-expression data associated with non-neuronal and non-neuroendocrine tumors. Differential treatment and/or treatment recommendations may be provided based on the classification. First-line checkpoint blockade therapy may be used or recommended when a tumor is identified as being non-neurally related, and a combination therapy (e.g., initial chemotherapy and subsequent checkpoint blockade therapy) may be used or recommended when a tumor is identified as being neurally related.

Abstract: The present invention provides diagnostic methods, therapeutic methods, and compositions for the treatment of cancer (e.g., a bladder cancer (e.g., UC, e.g., mUC), a kidney cancer, a lung cancer, a liver cancer, an ovarian cancer, a pancreatic cancer, a colorectal cancer, or a breast cancer). The invention is based, at least in part, on the discovery that expression levels of one or more biomarkers described herein in a sample from an individual having cancer can be used in methods of identifying an individual having a cancer who may benefit with an anti-cancer therapy that includes an immunotherapy (e.g., a PD-L1 axis binding antagonist such as an anti-PD-L1 antibody (e.g., atezolizumab)) and a suppressive stromal antagonist (e.g.