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.

Neoadjuvant immunotherapy plus chemotherapy improves event-free survival (EFS) and pathologic complete response (0% residual viable tumor (RVT) in primary tumor (PT) and lymph nodes (LNs)), and is approved for treatment of resectable lung cancer. Pathologic response assessment after neoadjuvant therapy is the potential analog to radiographic response for advanced disease. However, %RVT thresholds beyond pathologic complete response and major pathologic response (≤10% RVT) have not been explored. Pathologic response was prospectively assessed in the randomized, phase 3 CheckMate 816 trial (NCT02998528), which evaluated neoadjuvant nivolumab (anti-programmed death protein 1) plus chemotherapy in patients with resectable lung cancer. RVT, regression and necrosis were quantified (0–100%) in PT and LNs using a pan-tumor scoring system and tested for association with EFS in a prespecified exploratory analysis. Regardless of LN involvement, EFS improved with 0% versus >0% RVT-PT (hazard ratio = 0.18). RVT-PT predicted EFS for nivolumab plus chemotherapy (area under the curve = 0.74); 2-year EFS rates were 90%, 60%, 57% and 39% for patients with 0–5%, >5–30%, >30–80% and >80% RVT, respectively. Each 1% RVT associated with a 0.017 hazard ratio increase for EFS. Combining pathologic response from PT and LNs helped differentiate outcomes. When compared with radiographic response and circulating tumor DNA clearance, %RVT best approximated EFS. These findings support pathologic response as an emerging survival surrogate. Further assessment of the full spectrum of %RVT in lung cancer and other tumor types is warranted. ClinicalTrials.gov registration: NCT02998528.

MicroRNAs are short RNAs that serve as regulators of gene expression and are essential components of normal development as well as modulators of disease. MicroRNAs generally act cell-autonomously, and thus their localization to specific cell types is needed to guide our understanding of microRNA activity. Current tissue-level data have caused considerable confusion, and comprehensive cell-level data do not yet exist. Here, we establish the landscape of human cell-specific microRNA expression. This project evaluated 8 billion small RNA-seq reads from 46 primary cell types, 42 cancer or immortalized cell lines, and 26 tissues. It identified both specific and ubiquitous patterns of expression that strongly correlate with adjacent superenhancer activity. Analysis of unaligned RNA reads uncovered 207 unknown minor strand (passenger) microRNAs of known microRNA loci and 495 novel putative microRNA loci. Although cancer cell lines generally recapitulated the expression patterns of matched primary cells, their isomiR sequence families exhibited increased disorder, suggesting DROSHA- and DICER1-dependent microRNA processing variability. Cell-specific patterns of microRNA expression were used to de-convolute variable cellular composition of colon and adipose tissue samples, highlighting one use of these cell-specific microRNA expression data. Characterization of cellular microRNA expression across a wide variety of cell types provides a new understanding of this critical regulatory RNA species.

Introduction: Randomized controlled trials of platinum-based neoadjuvant chemotherapy (NAC) for bladder cancer have shown that patients who achieve a pathologic response to NAC exhibit 5 y survival rates of approximately 80–90% while NAC resistant (NR) cases exhibit 5 y survival rates of approximately 30–40%. These findings highlight the need to predict who will benefit from conventional NAC and the need for plausible alternatives. Methods: The pre-treatment biopsy tissues from a cohort of 41 patients with muscle invasive bladder who were treated with NAC were incorporated in tissue microarray and immunohistochemistry for PD-L1, CD8, and FOXP3 was performed. Percentage of PD-L1 positive tumor cells was measured. Tumor-infiltrating lymphocytes (TIL) densities, along with CD8 and Treg-specific TILs, were measured. Results: TIL density was strongly correlated with tumor PD-L1 expression, consistent with the mechanism of adaptive immune resistance in bladder cancer. Tumor cell PD-L1 expression was not a significant predictor of response. Neither was the CD8 nor Treg TIL density associated with response. Intriguingly though, the ratio of CD8 to Treg TIL densities was strongly associated with response (p = 0.0003), supporting the hypothesis that the immune system plays a role in the response of bladder cancer to chemotherapy. Discussion: To our knowledge, this is the first report in bladder cancer showing that the CD8 to Treg TIL density in the pre-treatment tissues is predictive for conventional NAC response. These findings warrant further investigations to both better characterize this association in larger cohorts and begin to elucidate the underlying mechanism(s) of this phenomenon.

In addition to its role as a TB vaccine, BCG has been shown to elicit heterologous protection against many other pathogens including viruses through a process termed trained immunity. Despite its potential as a broadly protective vaccine, little has been done to determine if BCG-mediated trained immunity levels can be optimized. Here we re-engineer BCG to express high levels of c-di-AMP, a PAMP recognized by stimulator of interferon genes (STING). We find that BCG overexpressing c-di-AMP elicits more potent signatures of trained immunity including higher pro-inflammatory cytokine responses, greater myeloid cell reprogramming toward inflammatory and activated states, and enhances epigenetic and metabolomic changes. In a model of bladder cancer, we also show that re-engineered BCG induces trained immunity and improved functionality. These results indicate that trained immunity levels and antitumor efficacy may be increased by modifying BCG to express higher levels of key PAMP molecules.

MiRNAs play important roles in the regulation of gene expression. The rapidly developing field of microRNA sequencing (miRNA-seq; small RNA-seq) needs comprehensive, robust, user-friendly and standardized bioinformatics tools to analyze these large datasets. We present miRge 2.0, in which multiple enhancements were made towards these goals. miRge 2.0 has become more comprehensive with increased functionality including a novel miRNA detection method, A-to-I editing analysis, integrated standardized GFF3 isomiR reporting, and improved alignment to miRNAs. The novel miRNA detection method uniquely uses both miRNA hairpin sequence structure and composition of isomiRs resulting in higher specificity for potential miRNA identification. Using known miRNA data, our support vector machine (SVM) model predicted miRNAs with an average Matthews correlation coefficient (MCC) of 0.939 over 32 human cell datasets and outperformed miRDeep2 and miRAnalyzer regarding phylogenetic conservation. The A-to-I editing detection strongly correlated with a reference dataset with adjusted R2 = 0.96. miRge 2.0 is the most up-to-date aligner with custom libraries to both miRBase v22 and MirGeneDB v2.0 for 6 species: human, mouse, rat, fruit fly, nematode and zebrafish; and has a tool to create custom libraries. For user-friendliness, miRge 2.0 is incorporated into bcbio-nextgen and implementable through Bioconda. miRge 2.0 is a redesigned, leading miRNA RNA-seq aligner with several improvements and novel utilities. miRge 2.0 is freely available at: https://github.com/mhalushka/miRge .