Abstract: The HIV surface contains viral glycoproteins (Env) that facilitate virus attachment and entry into host cells. In addition to Env, a broad range of human proteins are also present on the HIV surface, and these human proteins can also influence virus attachment and infectivity. We recently discovered many new human proteins present on HIV particles, and these data highlight gaps in our current knowledge and understanding of HIV biology. To better characterize the proteins present on HIV and their relevance to virus biology, we are pioneering new experimental approaches with Flow Virometry (FV) techniques. Recently, we have made new advances in our ability to perform multicolor immunophenotyping of virus particles, and the ability to stain capsid protein (Gag p24) inside virus particles, collectively enabling us to better discern heterogeneity in virus particles, and distinguish viruses from extracellular vesicles. We are also applying new experimental approaches for understanding HIV reservoirs, with data demonstrating highly distinct proteomic profiles of virus particles produced in different cell types. These data suggest that virions emerge from infected cells with a distinct protein profile that can inform on their original cellular source, and this knowledge may improve our understanding of viral reservoirs in people living with HIV.

Abstract: The repair path of DNA double strand breaks (DSBs) is determined by nuclear process context and the DNA Damage Response (DDR) to suppress chromosome translocations. Although primary DSB repair mechanisms like nonhomologous end joining and homologous recombination are well established, their disrepair and the contribution from alternative DSB repair mechanisms remain to be elucidated, limiting our ability to develop specific and potent cell-based therapies. In this seminar, I will discuss primary and secondary DNA end joining mechanisms, influenced by the DDR, in cycling and quiescent cells from the perspective of genome-wide junction capture studies of physiologic and engineered DSBs. Additionally, from parallel joint capture screens, I will describe several inhibitor classes and gene deficiencies with distinct roles in suppressing V(D)J recombination while separately promoting translocations between Cas9 DSBs. This latter part will highlight a speculative model for how the DDR promotes efficient single DSB rejoining.

Abstract: Transcriptomic sequencing is revolutionizing our exploration of Earth’s RNA virome. Yet current sequence-comparison methods fail to identify highly divergent viruses, and do not scale to the petabytes of available data. Collectively, the global research community has generated >60 petabases (6x1016 nt) of freely-avaialble sequencing data from 27+ million biological samples. These datasets encompass all continents, oceans, thousands of animals, plant, fungal, and microbial species, and are valued at $3.6-14.9 billion dollars in direct sequencing cost. To explore the total diversity of RNA viruses on Earth, we developed a cloud-based sequence alignment platform called Serratus (www.serratus.io) capable of petabase-scale sequence analysis. At the end of 2020, there were 15,000 known RNA viruses. We analyzed all 7.4 million public RNA sequencing datasets for the RNA viral hallmark gene, RNA-dependent RNA polymerase. This led us to identify >500,000 novel RNA virus species, which is over an order of magnitude more RNA viruses than previously known; including nine new species of highly divergent Coronaviruses. The Serratus platform has unlocked planetary-scale informatics, providing us with an unprecedented capacity to describe virus evolution and ecology. While exploring ultra-divergent RNA viruses, we uncovered a new phylum of globally distributed circular RNA viruses which share genomic features with viroids, and thus possibly represent a primordial transition from RNA world viroids, to protein-coding RNA viruses. Applying the platform to discover viroids, we have uncovered tens of thousands of naturally occurring ribozyme sequences specialized towards specific ecological niches, including the recent characterization of a new type of entity in the human microbiome, Obelisks. The next decade of virology will be illuminated by big-data analytics and requires a shift in our conceptualization of what is possible in virus discovery, and effective pandemic surveillance. What we learn today from a mere 26 million sequencing datasets will lay the foundations for how we will functionalize the expected 100+ million virus species we project to discover by 2030.

Abstract: Biomolecular condensates concentrate biomolecules in the cell, often formed by the physical process of liquid-liquid phase separation. By enriching specific biomolecules, biomolecular condensates can increase or decrease biochemical reactions, buffer protein concentrations, sense environmental changes, and provide mechanical forces. To study condensate function, we employ proximity-based labeling methods (BioID or APEX) to determine condensate proteomes. Biomolecular condensates are dynamic and metastable. Aberrant phase transition underlies the pathogenesis of neurodegenerative diseases. We apply quantitative mass spectrometry to measure changes in proximal interactions associated with condensate formation and aberrant phase transition. We focus our studies on cytosolic biomolecular condensates formed during stress, called stress granules (SGs) and TDP-43 (TARDBP) condensates.