Publications

A mechanistic understanding of the SARS-CoV-2 viral replication cycle is essential to develop new therapies for the COVID-19 global health crisis. In this study, we show that the SARS-CoV-2 nucleocapsid protein (N-protein) undergoes liquid-liquid phase separation (LLPS) with the viral genome, and propose a model of viral packaging through LLPS. N-protein condenses with specific RNA sequences in the first 1000 nts (5’-End) under physiological conditions and is enhanced at human upper airway temperatures. N-protein condensates exclude non-packaged RNA sequences. We comprehensively map sites bound by N-protein in the 5’-End and find preferences for single-stranded RNA flanked by stable structured elements. Liquid-like N-protein condensates form in mammalian cells in a concentration-dependent manner and can be altered by small molecules. Condensation of N-protein is sequence and structure specific, sensitive to human body temperature, and manipulatable with small molecules thus presenting screenable processes for identifying antiviral compounds effective against SARS-CoV-2.

Evolved stars near the tip of the red giant branch (TRGB) show solar-like oscillations with periods spanning hours to months and amplitudes ranging from ∼1 mmag to ∼100 mmag. The systematic detection of the resulting photometric variations with ground-based telescopes would enable the application of asteroseismology to a much larger and more distant sample of stars than is currently accessible with space-based telescopes such as \textit{Kepler} or the ongoing Transiting Exoplanet Survey Satellite (\textit{TESS}) mission. We present an asteroseismic analysis of 493 M giants using data from two ground-based surveys: the Asteroid Terrestrial-impact Last Alert System (ATLAS) and the All-Sky Automated Survey for Supernovae (ASAS-SN). By comparing the extracted frequencies with constraints from \textit{Kepler}, the Sloan Digital Sky Survey Apache Point Observatory Galaxy Evolution Experiment (APOGEE), and Gaia we demonstrate that ground-based transient surveys allow accurate distance measurements to oscillating M giants with a precision of ∼15%. Using stellar population synthesis models we predict that ATLAS and ASAS-SN can provide asteroseismic distances to ∼2×106 galactic M giants out to typical distances of 20−50kpc, vastly improving the reach of Gaia and providing critical constraints for Galactic archaeology and galactic dynamics.

The large number of available sequences and the diversity of protein functions challenge current experimental and computational approaches to determining and predicting protein function. We present a deep learning Graph Convolutional Network (GCN) for predicting protein functions and concurrently identifying functionally important residues. This model is initially trained using experimentally determined structures from the Protein Data Bank (PDB) but has significant de-noising capability, with only a minor drop in performance observed when structure predictions are used. We take advantage of this denoising property to train the model on > 200,000 protein structures, including many homology-predicted structures, greatly expanding the reach and applications of the method. Our model learns general structure-function relationships by robustly predicting functions of proteins with ≤ 40% sequence identity to the training set. We show that our GCN architecture predicts functions more accurately than Convolutional Neural Networks trained on sequence data alone and previous competing methods. Using class activation mapping, we automatically identify structural regions at the residue-level that lead to each function prediction for every confidently predicted protein, advancing site-specific function prediction. We use our method to annotate PDB and SWISS-MODEL proteins, making several new confident function predictions spanning both fold and function classifications.

We present a neural network architecture that is fully equivariant with respect to transformations under the Lorentz group, a fundamental symmetry of space and time in physics. The architecture is based on the theory of the finite-dimensional representations of the Lorentz group and the equivariant nonlinearity involves the tensor product. For classification tasks in particle physics, we show that such an equivariant architecture leads to drastically simpler models that have relatively few learnable parameters and are much more physically interpretable than leading approaches that use CNNs and point cloud approaches. The performance of the network is tested on a public classification dataset [https://zenodo.org/record/2603256] for tagging top quark decays given energy-momenta of jet constituents produced in proton-proton collisions.

The impressive pace of advance of quantum technology calls for robust and scalable techniques for the characterization and validation of quantum hardware. Quantum process tomography, the reconstruction of an unknown quantum channel from measurement data, remains the quintessential primitive to completely characterize quantum devices. However, due to the exponential scaling of the required data and classical post-processing, its range of applicability is typically restricted to one- and two-qubit gates. Here, we present a new technique for performing quantum process tomography that addresses these issues by combining a tensor network representation of the channel with a data-driven optimization inspired by unsupervised machine learning. We demonstrate our technique through synthetically generated data for ideal one- and two-dimensional random quantum circuits of up to 10 qubits, and a noisy 5-qubit circuit, reaching process fidelities above 0.99 using only a limited set of single-qubit measurement samples and input states. Our results go far beyond state-of-the-art, providing a practical and timely tool for benchmarking quantum circuits in current and near-term quantum computers.

Small noncoding RNAs (sRNAs) are key regulators of bacterial gene expression. Through complementary base pairing, sRNAs affect mRNA stability and translation efficiency. Here, we describe a network inference approach designed to identify sRNA-mediated regulation of transcript levels. We use existing transcriptional data sets and prior knowledge to infer sRNA regulons using our network inference tool, the Inferelator. This approach produces genome-wide gene regulatory networks that include contributions by both transcription factors and sRNAs. We show the benefits of estimating and incorporating sRNA activities into network inference pipelines using available experimental data. We also demonstrate how these estimated sRNA regulatory activities can be mined to identify the experimental conditions where sRNAs are most active. We uncover 45 novel experimentally supported sRNAmRNA interactions in Escherichia coli, outperforming previous network-based efforts. Additionally, our pipeline complements sequence-based sRNA-mRNA interaction prediction methods by adding a data-driven filtering step. Finally, we show the general
applicability of our approach by identifying 24 novel, experimentally supported, sRNA-mRNA interactions in Pseudomonas aeruginosa, Staphylococcus aureus, and Bacillus subtilis. Overall, our strategy generates novel insights into the functional context of sRNA regulation in multiple bacterial species.

stream membership probabilities, resulting in the detection of more than 200 likely members. These data show the stream extends to more than three times the length shown in the discovery data. A spur to the main stream is also detected. The high resolution spectra allow us to resolve the stellar velocity dispersion, found to be 1.6±0.3 km/sec.

We present a fast, high-order accurate and adaptive boundary integral scheme for solving the Stokes equations in complex---possibly nonsmooth---geometries in two dimensions. The key ingredient is a set of panel quadrature rules capable of evaluating weakly-singular, nearly-singular and hyper-singular integrals to high accuracy. Near-singular integral evaluation, in particular, is done using an extension of the scheme developed in J.~Helsing and R.~Ojala, {\it J. Comput. Phys.} {\bf 227} (2008) 2899--2921. The boundary of the given geometry is ``panelized'' automatically to achieve user-prescribed precision. We show that this adaptive panel refinement procedure works well in practice even in the case of complex geometries with large number of corners. In one example, for instance, a model 2D vascular network with 378 corners required less than 200K discretization points to obtain a 9-digit solution accuracy.

The transport of deformable particles through porous media underlies a wealth of applications ranging from filtration to oil recovery to the transport and spreading of biological agents. Using direct numerical simulations, we analyze the dynamics of semiflexible polymers under the influence of an imposed flow in a structured two-dimensional lattice serving as an idealization of a porous medium. This problem has received much attention in the limit of reptation and for long-chain polymer molecules such as DNA that are transported through micropost arrays for electrophoretic chromatographic separation. In contrast to long entropic molecules, the dynamics of elastic polymers results from a combination of scattering with the obstacles and flow-induced buckling instabilities. We identify three dominant modes of transport that involve trapping, gliding and vaulting of the polymers around the obstacles, and we reveal their essential features using tools from dynamical systems theory. The interplay of these scattering dynamics with transport and deformations in the imposed flow results in the long-time asymptotic dispersion of the center of mass, which we quantify in terms of a hydrodynamic dispersion tensor. We then discuss a simple yet efficient chromatographic device that exploits the competition between different modes of transport to sort filaments in a dilute suspension according to their lengths.