SUMMARY: Diversity of the T-cell receptor (TCR) repertoire is central to adaptive immunity. The TCR is composed of α and β chains, encoded by the TRA and TRB genes, of which the variable regions determine antigen specificity. To generate novel biological insights into the complex functioning of immune cells, combined capture of variable regions and single-cell transcriptomes provides a compelling approach. Recent developments enable the enrichment of TRA and TRB variable regions from widely used technologies for 3'-based single-cell RNA-sequencing (scRNA-seq). However, a comprehensive computational pipeline to process TCR-enriched data from 3' scRNA-seq is not available. Here, we present an analysis pipeline to process TCR variable regions enriched from 3' scRNA-seq cDNA. The tool reports TRA and TRB nucleotide and amino acid sequences linked to cell barcodes, enabling the reconstruction of T-cell clonotypes with associated transcriptomes. We demonstrate the software using peripheral blood mononuclear cells from a healthy donor and detect TCR sequences in a high proportion of single T cells. Detection of TCR sequences is low in non-T-cell populations, demonstrating specificity. Finally, we show that TCR clones are larger in CD8 Memory T cells than in other T-cell types, indicating an association between T-cell clonotypes and differentiation states. AVAILABILITY AND IMPLEMENTATION: The Workflow for Association of T-cell receptors from 3' single-cell RNA-seq (WAT3R), including test data, is available on GitHub (https://github.com/mainciburu/WAT3R), Docker Hub (https://hub.docker.com/r/mainciburu/wat3r) and a workflow on the Terra platform (https://app.terra.bio). The test dataset is available on GEO (accession number GSE195956). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: This work addresses two common issues in building classification models for biological or medical studies: learning a sparse model, where only a subset of a large number of possible predictors is used, and training in the presence of missing data. This work focuses on supervised generative binary classification models, specifically linear discriminant analysis (LDA). The parameters are determined using an expectation maximization algorithm to both address missing data and introduce priors to promote sparsity. The proposed algorithm, expectation-maximization sparse discriminant analysis (EM-SDA), produces a sparse LDA model for datasets with and without missing data. RESULTS: EM-SDA is tested via simulations and case studies. In the simulations, EM-SDA is compared with nearest shrunken centroids (NSCs) and sparse discriminant analysis (SDA) with k-nearest neighbors for imputation for varying mechanism and amount of missing data. In three case studies using published biomedical data, the results are compared with NSC and SDA models with four different types of imputation, all of which are common approaches in the field. EM-SDA is more accurate and sparse than competing methods both with and without missing data in most of the experiments. Furthermore, the EM-SDA results are mostly consistent between the missing and full cases. Biological relevance of the resulting models, as quantified via a literature search, is also presented. AVAILABILITY AND IMPLEMENTATION: A Matlab implementation published under GNU GPL v.3 license is available at http://web.mit.edu/braatzgroup/links.html . CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Recently, there have been significant advances in the fabrication and demonstration of individual molecular electronic wires and diode switches. This paper reviews those developments and shows how demonstrated molecular devices might be combined to design molecular-scale electronic digital computer logic. The design for the demonstrated rectifying molecular diode switches is refined and made more compatible with the demonstrated wires through the introduction of intramolecular dopant groups chemically bonded to modified molecular wires. Quantum mechanical calculations are performed to characterize some of the electrical properties of the proposed molecular diode switches. Explicit structural designs are displayed for AND, OR, and XOR gates that are built from molecular wires and molecular diode switches. The diode-based molecular electronic logic gates are combined to produce a design for a molecular-scale electronic half adder and a molecular-scale electronic full adder. These designs correspond to conductive monomolecular circuit structures that would be one million times smaller in area than the corresponding micron-scale digital logic circuits fabricated on conventional solid-state semiconductor computer chips. It appears likely that these nanometer-scale molecular electronic logic circuits could be fabricated and tested in the foreseeable future. At the very least, such molecular circuit designs constitute an exploration of the ultimate limits of electronic computer circuit miniaturization.

This paper provides an overview of research developments toward nanometer-scale electronic switching devices for use in building ultra-densely integrated electronic computers. Specifically, two classes of alternatives to the field-effect transistor are considered: (1) quantum-effect and single-electron solid-state devices and (2) molecular electronic devices. A taxonomy of devices in each class is provided, operational principles are described and compared for the various types of devices, and the literature about each is surveyed. This information is presented in nonmathematical terms intended for a general, technically interested readership.