PandoraGAN: Generating antiviral peptides using Generative Adversarial Network

MotivationThe continuous increase in pathogenic viruses and the intensive laboratory research for development of novel antiviral therapies often poses challenge in terms of cost and time efficient drug design. This accelerates research for alternate drug candidates and contributes to recent rise in research of antiviral peptides against many of the viruses. With limited information regarding these peptides and their activity, modifying the existing peptide backbone or developing a novel peptide is very time consuming and a tedious process. Advanced deep learning approaches such as generative adversarial networks (GAN) can be helpful for wet lab scientist to screen potential antiviral candidates of interest and expedite the initial stage of peptide drug development. To our knowledge this is the first ever use of GAN models for antiviral peptides across the viral spectrum.ResultsIn this study, we develop PandoraGAN that utilizes GAN to design bio active antiviral peptides. Available antiviral peptide data was manually curated for preparing highly active peptides data set to include peptides with lower IC50 values. We further validated the generated sequences comparing the physico-chemical properties of generated antiviral peptides with manually curated highly active training data.AvailabilityAntiviral sequences generated by PandoraGAN are available on PandoraGAN server. https://pandora-gan.herokuapp.com/. The code is available at https://gitlab.com/shraddha.surana/antiviral-peptide-predictions-using-ganContactparora@thoughtworks.com

A Broad-Spectrum Antiviral Peptide Blocks Infection of Viruses by Binding to Phosphatidylserine in the Viral Envelope

The ongoing threat of viral infections and the emergence of antiviral drug resistance warrants a ceaseless search for new antiviral compounds. Broadly-inhibiting compounds that act on elements shared by many viruses are promising antiviral candidates. Here, we identify a peptide derived from the cowpox virus protein CPXV012 as a broad-spectrum antiviral peptide. We found that CPXV012 peptide hampers infection by a multitude of clinically and economically important enveloped viruses, including poxviruses, herpes simplex virus-1, hepatitis B virus, HIV-1, and Rift Valley fever virus. Infections with non-enveloped viruses such as Coxsackie B3 virus and adenovirus are not affected. The results furthermore suggest that viral particles are neutralized by direct interactions with CPXV012 peptide and that this cationic peptide may specifically bind to and disrupt membranes composed of the anionic phospholipid phosphatidylserine, an important component of many viral membranes. The combined results strongly suggest that CPXV012 peptide inhibits virus infections by direct interactions with phosphatidylserine in the viral envelope. These results reiterate the potential of cationic peptides as broadly-acting virus inhibitors.

An Antiviral Peptide from Alopecosa nagpag Spider Targets NS2B–NS3 Protease of Flaviviruses

Flaviviruses are single-stranded RNA viruses predominantly transmitted by the widely distributed Aedes mosquitoes in nature. As important human pathogens, the geographic reach of Flaviviruses and their threats to public health are increasing, but there is currently no approved specific drug for treatment. In recent years, the development of peptide antivirals has gained much attention. Natural host defense peptides which uniquely evolved to protect the hosts have been shown to have antiviral properties. In this study, we firstly collected the venom of the Alopecosa nagpag spider from Shangri-La County, Yunnan Province. A defense peptide named Av-LCTX-An1a (Antiviral-Lycotoxin-An1a) was identified from the spider venom, and its anti-dengue serotype-2 virus (DENV2) activity was verified in vitro. Moreover, a real-time fluorescence-based protease inhibition assay showed that An1a functions as a DENV2 NS2B–NS3 protease inhibitor. Furthermore, we also found that An1a restricts zika virus (ZIKV) infection by inhibiting the ZIKV NS2B–NS3 protease. Together, our findings not only demonstrate that An1a might be a candidate for anti-flavivirus drug but also indicate that spider venom is a potential resource library rich in antiviral precursor molecules.