Deep mutational learning predicts ACE2 binding and antibody escape to combinatorial mutations in the SARS-CoV-2 receptor-binding domain.

The continual evolution of SARS-CoV-2 and the emergence of variants that show resistance to vaccines and neutralizing antibodies threaten to prolong the COVID-19 pandemic. Selection and emergence of SARS-CoV-2 variants are driven in part by mutations within the viral spike protein and in particular the ACE2 receptor-binding domain (RBD), a primary target site for neutralizing antibodies. Here, we develop deep mutational learning (DML), a machine-learning-guided protein engineering technology, which is used to investigate a massive sequence space of combinatorial mutations, representing billions of RBD variants, by accurately predicting their impact on ACE2 binding and antibody escape. A highly diverse landscape of possible SARS-CoV-2 variants is identified that could emerge from a multitude of evolutionary trajectories. DML may be used for predictive profiling on current and prospective variants, including highly mutated variants such as Omicron, thus guiding the development of therapeutic antibody treatments and vaccines for COVID-19.

DeepSARS: simultaneous diagnostic detection and genomic surveillance of SARS-CoV-2.

BACKGROUND: The continued spread of SARS-CoV-2 and emergence of new variants with higher transmission rates and/or partial resistance to vaccines has further highlighted the need for large-scale testing and genomic surveillance. However, current diagnostic testing (e.g., PCR) and genomic surveillance methods (e.g., whole genome sequencing) are performed separately, thus limiting the detection and tracing of SARS-CoV-2 and emerging variants. RESULTS: Here, we developed DeepSARS, a high-throughput platform for simultaneous diagnostic detection and genomic surveillance of SARS-CoV-2 by the integration of molecular barcoding, targeted deep sequencing, and computational phylogenetics. DeepSARS enables highly sensitive viral detection, while also capturing genomic diversity and viral evolution. We show that DeepSARS can be rapidly adapted for identification of emerging variants, such as alpha, beta, gamma, and delta strains, and profile mutational changes at the population level. CONCLUSIONS: DeepSARS sets the foundation for quantitative diagnostics that capture viral evolution and diversity. DeepSARS uses molecular barcodes (BCs) and multiplexed targeted deep sequencing (NGS) to enable simultaneous diagnostic detection and genomic surveillance of SARS-CoV-2. Image was created using Biorender.com .

Proteomic Analysis Reveals a Role for RSK in p120-catenin Phosphorylation and Melanoma Cell-Cell Adhesion.

The RAS/mitogen-activated protein kinase (MAPK) signaling pathway regulates various biological functions, including cell survival, proliferation and migration. This pathway is frequently deregulated in cancer, including melanoma, which is the most aggressive form of skin cancer. RSK (p90 ribosomal S6 kinase) is a MAPK-activated protein kinase required for melanoma growth and proliferation, but relatively little is known about its function and the nature of its cellular partners. In this study, we used a proximity-based labeling approach to identify RSK proximity partners in cells. We identified many potential RSK-interacting proteins, including p120ctn (p120-catenin), which is an essential component of adherens junction (AJ). We found that RSK phosphorylates p120ctn on Ser320, which appears to be constitutively phosphorylated in melanoma cells. We also found that RSK inhibition increases melanoma cell-cell adhesion, suggesting that constitutive RAS/MAPK signaling negatively regulates AJ integrity. Together, our results indicate that RSK plays an important role in the regulation of melanoma cell-cell adhesion.

Translational control by oncogenic signaling pathways.

Messenger RNA (mRNA) translation is highly regulated in cells and plays an integral role in the overall process of gene expression. The initiation phase of translation is considered to be the most rate-limiting and is often targeted by oncogenic signaling pathways to promote global protein synthesis and the selective translation of tumor-promoting mRNAs. Translational control is a crucial component of cancer development as it allows cancer cells to adapt to the altered metabolism that is generally associated with the tumor state. The phosphoinositide 3-kinase (PI3K)/Akt and Ras/mitogen-activated protein kinase (MAPK) pathways are strongly implicated in cancer etiology, and they exert their biological effects by modulating both global and specific mRNA translation. In addition to having respective translational targets, these pathways also impinge on the mechanistic/mammalian target of rapamycin (mTOR), which acts as a critical signaling node linking nutrient sensing to the coordinated regulation of cellular metabolism. mTOR is best known as a central regulator of protein synthesis and has been implicated in an increasing number of pathological conditions, including cancer. In this article, we describe the current knowledge on the roles and regulation of mRNA translation by various oncogenic signaling pathways, as well as the relevance of these molecular mechanisms to human malignancies. This article is part of a Special Issue entitled: Translation and cancer.

Learning what not to select for in antibody drug discovery.

Identifying antibodies with high affinity and target specificity is crucial for drug discovery and development; however, filtering out antibody candidates with nonspecific or polyspecific binding profiles is also important. In this issue of Cell Reports Methods, Saksena et al. report a computational counterselection method combining deep sequencing and machine learning for identifying nonspecific antibody candidates and demonstrate that it has advantages over more established molecular counterselection methods.

Applications of Machine and Deep Learning in Adaptive Immunity.

Adaptive immunity is mediated by lymphocyte B and T cells, which respectively express a vast and diverse repertoire of B cell and T cell receptors and, in conjunction with peptide antigen presentation through major histocompatibility complexes (MHCs), can recognize and respond to pathogens and diseased cells. In recent years, advances in deep sequencing have led to a massive increase in the amount of adaptive immune receptor repertoire data; additionally, proteomics techniques have led to a wealth of data on peptide-MHC presentation. These large-scale data sets are now making it possible to train machine and deep learning models, which can be used to identify complex and high-dimensional patterns in immune repertoires. This article introduces adaptive immune repertoires and machine and deep learning related to biological sequence data and then summarizes the many applications in this field, which span from predicting the immunological status of a host to the antigen specificity of individual receptors and the engineering of immunotherapeutics.

Flotillin-mediated endocytosis and ALIX-syntenin-1-mediated exocytosis protect the cell membrane from damage caused by necroptosis.

Necroptosis is a form of regulated necrosis that is implicated in various human diseases including Alzheimer's disease. Necroptosis requires the translocation of the pseudokinase MLKL from the cytosol to the plasma membrane after its phosphorylation by the kinase RIPK3. Using protein cross-linking followed by affinity purification, we detected the lipid raft-associated proteins flotillin-1 and flotillin-2 and the ESCRT-associated proteins ALIX and syntenin-1 in membrane-localized MLKL immunoprecipitates. Phosphorylated MLKL was removed from membranes through either flotillin-mediated endocytosis followed by lysosomal degradation or ALIX-syntenin-1-mediated exocytosis. Thus, cells undergoing necroptosis need to overcome these independent suppressive mechanisms before plasma membrane disruption can occur.