Genome-wide Identification of Structure-Forming Repeats as Principal Sites of Fork Collapse upon ATR Inhibition.

DNA polymerase stalling activates the ATR checkpoint kinase, which in turn suppresses fork collapse and breakage. Herein, we describe use of ATR inhibition (ATRi) as a means to identify genomic sites of problematic DNA replication in murine and human cells. Over 500 high-resolution ATR-dependent sites were ascertained using two distinct methods: replication protein A (RPA)-chromatin immunoprecipitation (ChIP) and breaks identified by TdT labeling (BrITL). The genomic feature most strongly associated with ATR dependence was repetitive DNA that exhibited high structure-forming potential. Repeats most reliant on ATR for stability included structure-forming microsatellites, inverted retroelement repeats, and quasi-palindromic AT-rich repeats. Notably, these distinct categories of repeats differed in the structures they formed and their ability to stimulate RPA accumulation and breakage, implying that the causes and character of replication fork collapse under ATR inhibition can vary in a DNA-structure-specific manner. Collectively, these studies identify key sources of endogenous replication stress that rely on ATR for stability.

In Vivo Delivery of a DNA-Encoded Monoclonal Antibody Protects Non-human Primates against Zika Virus.

Zika virus (ZIKV) infection is endemic to several world regions, and many others are at high risk for seasonal outbreaks. Synthetic DNA-encoded monoclonal antibody (DMAb) is an approach that enables in vivo delivery of highly potent mAbs to control infections. We engineered DMAb-ZK190, encoding the mAb ZK190 neutralizing antibody, which targets the ZIKV E protein DIII domain. In vivo-delivered DMAb-ZK190 achieved expression levels persisting >10 weeks in mice and >3 weeks in non-human primate (NHPs), which is protective against ZIKV infectious challenge. This study is the first demonstration of infectious disease control in NHPs following in vivo delivery of a nucleic acid-encoded antibody, supporting the importance of this new platform.

Building trust in deep learning-based immune response predictors with interpretable explanations.

The ability to predict whether a peptide will get presented on Major Histocompatibility Complex (MHC) class I molecules has profound implications in designing vaccines. Numerous deep learning-based predictors for peptide presentation on MHC class I molecules exist with high levels of accuracy. However, these MHC class I predictors are treated as black-box functions, providing little insight into their decision making. To build turst in these predictors, it is crucial to understand the rationale behind their decisions with human-interpretable explanations. We present MHCXAI, eXplainable AI (XAI) techniques to help interpret the outputs from MHC class I predictors in terms of input peptide features. In our experiments, we explain the outputs of four state-of-the-art MHC class I predictors over a large dataset of peptides and MHC alleles. Additionally, we evaluate the reliability of the explanations by comparing against ground truth and checking their robustness. MHCXAI seeks to increase understanding of deep learning-based predictors in the immune response domain and build trust with validated explanations.

A web-based automated machine learning platform to analyze liquid biopsy data.

Liquid biopsy (LB) technologies continue to improve in sensitivity, specificity, and multiplexing and can measure an ever growing library of disease biomarkers. However, clinical interpretation of the increasingly large sets of data these technologies generate remains a challenge. Machine learning is a popular approach to discover and detect signatures of disease. However, limited machine learning expertise in the LB field has kept the discipline from fully leveraging these tools and risks improper analyses and irreproducible results. In this paper, we develop a web-based automated machine learning tool tailored specifically for LB, where machine learning models can be built without the user's input. We also incorporate a differential privacy algorithm, designed to limit the effects of overfitting that can arise from users iteratively developing a panel with feedback from our platform. We validate our approach by performing a meta-analysis on 11 published LB datasets, and found that we had similar or better performance compared to those reported in the literature. Moreover, we show that our platform's performance improved when incorporating information from prior LB datasets, suggesting that this approach can continue to improve with increased access to LB data. Finally, we show that by using our platform the results achieved in the literature can be matched using 40% of the number of subjects in the training set, potentially reducing study cost and time. This self-improving and overfitting-resistant automatic machine learning platform provides a new standard that can be used to validate machine learning works in the LB field.

Synthetic nucleic acid antibody prophylaxis confers rapid and durable protective immunity against Zika virus challenge.

Significant concerns have arisen over the past 3 y from the increased global spread of the mosquito-borne flavivirus, Zika. Accompanying this spread has been an increase in cases of the devastating birth defect microcephaly as well as of Guillain-Barré syndrome in adults in many affected countries. Currently there is no vaccine or therapy for this infection; however, we sought to develop a combination approach that provides more rapid and durable protection than traditional vaccination alone. A novel immune-based prophylaxis/therapy strategy entailing the facilitated delivery of a synthetic DNA consensus prME vaccine along with DNA-encoded anti-ZIKV envelope monoclonal antibodies (dMAb) were developed and evaluated for antiviral efficacy. This immediate and persistent protection strategy confers the ability to overcome shortcomings inherent with conventional active vaccination or passive immunotherapy. A collection of novel dMAbs were developed which were potent against ZIKV and could be expressed in serum within 24-48 h of in vivo administration. The DNA vaccine, from a previous development, was potent after adaptive immunity was developed, protecting against infection, brain and testes pathology in relevant mouse challenge models and in an NHP challenge. Delivery of potent dMAbs protected mice from the same murine viral challenge within days of delivery. Combined injection of dMAb and the DNA vaccine afforded rapid and long-lived protection in this challenge model, providing an important demonstration of the advantage of this synergistic approach to pandemic outbreaks.

Protective immunity by an engineered DNA vaccine for Mayaro virus.

Mayaro virus (MAYV) of the genus alphavirus is a mosquito-transmitted emerging infectious disease that causes an acute febrile illness, rash, headaches, and nausea that may turn into incapacitating, persistent arthralgias in some victims. Since its discovery in Trinidad in 1954, cases of MAYV infection have largely been confined there and to the northern countries of South America, but recently, MAYV cases have been reported in some island nations in the Caribbean Sea. Accompanying these reports is evidence that new vectors, including Aedes spp. mosquitos, recently implicated in the global spread of Zika and chikungunya viruses, are competent for MAYV transmission, which, if true, could facilitate the spread of MAYV beyond its current range. Despite its status as an emerging virus, there are no licensed vaccines to prevent MAYV infection nor therapeutics to treat it. Here, we describe the development and testing of a novel DNA vaccine, scMAYV-E, that encodes a synthetically-designed consensus MAYV envelope sequence. In vivo electroporation-enhanced immunization of mice with this vaccine induced potent humoral responses including neutralizing antibodies as well as robust T-cell responses to multiple epitopes in the MAYV envelope. Importantly, these scMAYV-E-induced immune responses protected susceptible mice from morbidity and mortality following a MAYV challenge.