Analyses

Zhang and colleagues analyze single-cell data from patients treated with immunotherapy in five cancer types and find that CXCL13-expressing subsets are implicated in response to treatment in the CD8⁺ and CD4⁺ T cell compartments.

Bao et. al. develop the algorithm Starfish, to identify six signatures of complex genomic rearrangements in human cancer genomics datasets, including a pattern called hourglass chromothripsis which is prominent in prostate cancer.

Zhang and colleagues perform systematic multiomics and functional integration of cell-surface proteins and develop a comprehensive catalog of cell-surface actionable targets across cancer, with a practical web platform to explore these data.

Ma et al. apply few-shot learning to train a neural network model on cell-line drug-response data, and they subsequently transfer it to distinct biological contexts including different tissues and patient-derived tumor cells and xenografts.

Degasperi et al. introduce a practical framework and Signal, an online tool, to analyze mutational signatures. They find evidence of tissue-specific variability in mutational signatures, which may impact tumor classification and clinical application.

Martínez-Jiménez et al. report how disruption of the ubiquitin–proteasome system affects cancer, estimating that >10% of driver mutations involve alterations in genes relevant in ubiquitin-mediated proteolysis, including E3 ligases and their targets.