Alexander S. Baras

Abstract The AACR Project GENIE is an international data-sharing consortium focused on generating an evidence base for precision cancer medicine by integrating clinical-grade cancer genomic data with clinical outcome data for tens of thousands of cancer patients treated at multiple institutions worldwide. In conjunction with the first public data release from approximately 19,000 samples, we describe the goals, structure, and data standards of the consortium and report conclusions from high-level analysis of the initial phase of genomic data. We also provide examples of the clinical utility of GENIE data, such as an estimate of clinical actionability across multiple cancer types (>30%) and prediction of accrual rates to the NCI-MATCH trial that accurately reflect recently reported actual match rates. The GENIE database is expected to grow to >100,000 samples within 5 years and should serve as a powerful tool for precision cancer medicine. Significance: The AACR Project GENIE aims to catalyze sharing of integrated genomic and clinical datasets across multiple institutions worldwide, and thereby enable precision cancer medicine research, including the identification of novel therapeutic targets, design of biomarker-driven clinical trials, and identification of genomic determinants of response to therapy. Cancer Discov; 7(8); 818–31. ©2017 AACR. See related commentary by Litchfield et al., p. 796. This article is highlighted in the In This Issue feature, p. 783

Deep learning algorithms have been utilized to achieve enhanced performance in pattern-recognition tasks. The ability to learn complex patterns in data has tremendous implications in immunogenomics. T-cell receptor (TCR) sequencing assesses the diversity of the adaptive immune system and allows for modeling its sequence determinants of antigenicity. We present DeepTCR, a suite of unsupervised and supervised deep learning methods able to model highly complex TCR sequencing data by learning a joint representation of a TCR by its CDR3 sequences and V/D/J gene usage. We demonstrate the utility of deep learning to provide an improved 'featurization' of the TCR across multiple human and murine datasets, including improved classification of antigen-specific TCRs and extraction of antigen-specific TCRs from noisy single-cell RNA-Seq and T-cell culture-based assays. Our results highlight the flexibility and capacity for deep neural networks to extract meaningful information from complex immunogenomic data for both descriptive and predictive purposes.

Astronomy accelerates tumor imaging Immunohistochemical stains for individual markers revolutionized diagnostic pathology decades ago but cannot capture enough information to accurately predict response to immunotherapy. Newer multiplex immunofluorescent technologies provide the potential to visualize the expression patterns of many functionally relevant molecules but present numerous challenges in accurate image analysis and data handling, particularly over large tumor areas. Drawing from the field of astronomy, in which petabytes of imaging data are routinely analyzed across a wide spectral range, Berry et al. developed a platform for multispectral imaging of whole-tumor sections with high-fidelity single-cell resolution. The resultant AstroPath platform was used to develop a multiplex immunofluorescent assay highly predictive of responses and outcomes for melanoma patients receiving immunotherapy. Science , aba2609, this issue p. eaba2609