Graphical Abstract Highlights d The exRNA Atlas provides access to human exRNA profiles and web-accessible tools d Atlas analysis reveals six exRNA cargo types present across five human biofluids d Five of the cargo types associate with specific vesicular and non-vesicular carriers d These findings and resources empower studies of extracellular RNA communication An extracellular RNA atlas from five human biofluids (serum, plasma, cerebrospinal fluid, saliva, and urine) reveals six extracellular RNA cargo types, including both vesicular and nonvesicular carriers. SUMMARY To develop a map of cell-cell communication mediated by extracellular RNA (exRNA), the NIH Extracellular RNA Communication Consortium created the exRNA Atlas resource (https://exrna-atlas.org). The Atlas version 4P1 hosts 5,309 exRNA-seq and exRNA qPCR profiles from 19 studies and a suite of analysis and visualization tools. To analyze variation between profiles, we apply computational deconvolution. The analysis leads to a model with six exRNA cargo types (CT1, CT2, CT3A, CT3B, CT3C, CT4), each detectable in multiple biofluids (serum, plasma, CSF, saliva, urine). Five of the cargo types associate with known vesicular and non-vesicular (lipoprotein and ribonucleoprotein) exRNA carriers. To validate utility of this model, we re-analyze an exercise response study by deconvolution to identify physiologically relevant response pathways that were not detected previously.To enable wide application of this model, as part of the exRNA Atlas resource, we provide tools for deconvolution and analysis of user-provided case-control studies.

The effectiveness of most cancer targeted therapies is short-lived. Tumors often develop resistance that might be overcome with drug combinations. However, the number of possible combinations is vast, necessitating data-driven approaches to find optimal patient-specific treatments. Here we report AstraZeneca’s large drug combination dataset, consisting of 11,576 experiments from 910 combinations across 85 molecularly characterized cancer cell lines, and results of a DREAM Challenge to evaluate computational strategies for predicting synergistic drug pairs and biomarkers. 160 teams participated to provide a comprehensive methodological development and benchmarking. Winning methods incorporate prior knowledge of drug-target interactions. Synergy is predicted with an accuracy matching biological replicates for >60% of combinations. However, 20% of drug combinations are poorly predicted by all methods. Genomic rationale for synergy predictions are identified, including ADAM17 inhibitor antagonism when combined with PIK3CB/D inhibition contrasting to synergy when combined with other PI3K-pathway inhibitors in PIK3CA mutant cells.

Many bioinformatics methods have been proposed for reducing the complexity of large gene or protein networks into relevant subnetworks or modules. Yet, how such methods compare to each other in terms of their ability to identify disease-relevant modules in different types of network remains poorly understood. We launched the ‘Disease Module Identification DREAM Challenge’, an open competition to comprehensively assess module identification methods across diverse protein–protein interaction, signaling, gene co-expression, homology and cancer-gene networks. Predicted network modules were tested for association with complex traits and diseases using a unique collection of 180 genome-wide association studies. Our robust assessment of 75 module identification methods reveals top-performing algorithms, which recover complementary trait-associated modules. We find that most of these modules correspond to core disease-relevant pathways, which often comprise therapeutic targets. This community challenge establishes biologically interpretable benchmarks, tools and guidelines for molecular network analysis to study human disease biology.

Clonal evolution of a tumor ecosystem depends on different selection pressures that are principally immune and treatment mediated. We integrate RNA-seq, DNA sequencing, TCRseq and SNP array data across multiple regions of liver cancer specimens to map spatio-temporal interactions between cancer and immune cells. We investigate how these interactions reflect intra-tumor heterogeneity (ITH) by correlating regional neo-epitope and viral antigen burden with the regional adaptive immune response. Regional expression of passenger mutations dominantly recruits adaptive responses as opposed to hepatitis B virus and cancer-testis antigens. We detect different clonal expansion of the adaptive immune system in distant regions of the same tumor. An ITH-based gene signature improves singlebiopsy patient survival predictions and an expression survey of 38,553 single cells across 7 regions of 2 patients further reveals heterogeneity in liver cancer. These data quantify transcriptomic ITH and how the different components of the HCC ecosystem interact during cancer evolution.

IMPORTANCE Mammography screening currently relies on subjective human interpretation. Artificial intelligence (AI) advances could be used to increase mammography screening accuracy by reducing missed cancers and false positives. OBJECTIVE To evaluate whether AI can overcome human mammography interpretation limitations with a rigorous, unbiased evaluation of machine learning algorithms. DESIGN, SETTING, AND PARTICIPANTS In this diagnostic accuracy study conducted between September 2016 and November 2017, an international, crowdsourced challenge was hosted to foster AI algorithm development focused on interpreting screening mammography. More than 1100 participants comprising 126 teams from 44 countries participated. Analysis began November 18, 2016. MAIN OUTCOMES AND MEASUREMENTS Algorithms used images alone (challenge 1) or combined images, previous examinations (if available), and clinical and demographic risk factor data (challenge 2) and output a score that translated to cancer yes/no within 12 months. Algorithm accuracy for breast cancer detection was evaluated using area under the curve and algorithm specificity compared with radiologists' specificity with radiologists' sensitivity set at 85.9% (United States) and 83.9% (Sweden). An ensemble method aggregating top-performing AI algorithms and radiologists' recall assessment was developed and evaluated. RESULTS Overall, 144 231 screening mammograms from 85 580 US women (952 cancer positive Յ12 months from screening) were used for algorithm training and validation. A second independent validation cohort included 166 578 examinations from 68 008 Swedish women (780 cancer positive). The top-performing algorithm achieved an area under the curve of 0.858 (United States) and 0.903 (Sweden) and 66.2% (United States) and 81.2% (Sweden) specificity at the radiologists' sensitivity, lower than community-practice radiologists' specificity of 90.5% (United States) and 98.5% (Sweden). Combining top-performing algorithms and US radiologist assessments resulted in a higher area under the curve of 0.942 and achieved a significantly improved specificity (92.0%) at the same sensitivity. CONCLUSIONS AND RELEVANCE While no single AI algorithm outperformed radiologists, an ensemble of AI algorithms combined with radiologist assessment in a single-reader screening environment improved overall accuracy. This study underscores the potential of using machine (continued)