Search for: All records

Sparse data structures like hash tables, trees, or compressed tensors are ubiquitous, but operations on these structures are expensive and inefficient on current systems. Prior work has proposed hardware acceleration for these operations, but these techniques have two key shortcomings: they limit the types of data structures they support, and they focus on reads but do not support fine-grained updates to these structures. We present Terminus, a programmable accelerator for read and update operations on sparse data structures. Terminus extends each general-purpose core with a programmable dataflow engine capable of accelerating a wide range of structures and operations. Terminus engines are flexible yet simple, as they focus on common operations and defer rare, complex ones to cores. Terminus features a simple concurrency control mechanism based on address ranges that enables safe updates while preserving parallelism. We evaluate Terminus on serial and parallel benchmarks on a wide range of sparse data structures. Terminus improves performance by gmean 7.4x over a CPU baseline, showing that Terminus can accelerate fine-grained reads and writes that were previously not possible in prior accelerators for sparse structures. 

Abstract Solute structure and its evolution in supersaturated aqueous solutions are key clues to understand Ostwald’s step rule. Here, we measure the structural evolution of solute molecules in highly supersaturated solutions of KH₂PO₄(KDP) and NH₄H₂PO₄(ADP) using a combination of electrostatic levitation and synchrotron X-ray scattering. The measurement reveals the existence of a solution-solution transition in KDP solution, caused by changing molecular symmetries and structural evolution of the solution with supersaturation. Moreover, we find that the molecular symmetry of H₂PO₄^-impacts on phase selection. These findings manifest that molecular symmetry and its structural evolution can govern the crystallization pathways in aqueous solutions, explaining the microscopic origin of Ostwald’s step rule. 

Two plants connected via a Cuscuta bridge exchange rapid systemic calcium, electric, and reactive oxygen species signals, suggesting that Cuscuta may have beneficial effects to host plants. 

Conjugating biomolecules, such as antibodies, to bioconjugate moieties on lipid surfaces is a powerful tool for engineering the surface of diverse biomaterials, including cells and nanoparticles. We developed supported lipid bilayers (SLBs) presenting well-defined spatial distributions of functional moieties as models for precisely engineered functional biomolecular-lipid surfaces. We used quartz crystal microbalance with dissipation (QCM-D) and atomic force microscopy (AFM) to determine how vesicles containing a mixture of 1,2-dipalmitoyl-sn-glycero-3-phosphatidylcholine (DPPC) and 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[azido(polyethylene glycol)-2000] (DSPE-PEG-N3) form SLBs as a function of the lipid phase transition temperature (Tm). Above the DPPC Tm, DPPC/DSPE-PEG-N3 vesicles form SLBs with functional azide moieties on SiO2 substrates via vesicle fusion. Below this Tm, DPPC/DSPE-PEG-N3 vesicles attach to SiO2 intact. Intact DPPC/DSPE-PEG-N3 vesicles on the SiO2 surfaces fuse and rupture to form SLBs when temperature is brought above the DPPC Tm. AFM studies show uniform and complete DPPC/DSPE-PEG-N3 SLB coverage of SiO2 surfaces for different DSPE-PEG-N3 concentrations. As the DSPE-PEG-N3 concentration increases from 0.01 to 6 mol%, the intermolecular spacing of DSPE-PEG-N3 in the SLBs decreases from 4.6 to 1.0 nm. The PEG moiety undergoes a mushroom to brush transition as DSPE-PEG-N3 concentration varies from 0.1 to 2.0 mol%. Via copper-free click reaction, IgG was conjugated to SLB surfaces with 4.6 nm or 1.3 nm inter-DSPE-PEG-N3 spacing. QCM-D and AFM data show; 1) uniform and complete IgG layers of similar mass and thickness on the two types of SLB; 2) a higher-viscosity/less rigid IgG layer on the SLB with 4.6 nm inter-DSPE-PEG-N3 spacing. Our studies provide a blueprint for SLBs modeling spatial control of functional macromolecules on lipid surfaces, including surfaces of lipid nanoparticles and cells. 

Cell identification is an important yet difficult process in data analysis of biological images. Previously, we developed an automated cell identification method called CRF_ID and demonstrated its high performance in C. elegans whole-brain images (Chaudhary et al, 2021). However, because the method was optimized for whole-brain imaging, comparable performance could not be guaranteed for application in commonly used C. elegans multi-cell images that display a subpopulation of cells. Here, we present an advance CRF_ID 2.0 that expands the generalizability of the method to multi-cell imaging beyond whole-brain imaging. To illustrate the application of the advance, we show the characterization of CRF_ID 2.0 in multi-cell imaging and cell-specific gene expression analysis in C. elegans. This work demonstrates that high accuracy automated cell annotation in multi-cell imaging can expedite cell identification and reduce its subjectivity in C. elegans and potentially other biological images of various origins. 

The COVID-19 pandemic has emphasized the urgency for effective antiviral therapies against SARS-CoV-2. Targeting the main protease (3CLpro) of the virus has emerged as a promising approach, and nirmatrelvir (PF-07321332), the active component of Pfizer’s oral drug Paxlovid, has demonstrated remarkable clinical efficacy. However, the emergence of resistance mutations poses a challenge to its continued success. In this study, we employed alchemical free energy perturbation (FEP) alanine scanning to identify nirmatrelvir-resistance mutations within SARS-CoV-2 3CLpro. FEP identified several mutations, which were validated through in vitro IC50 experiments and found to result in 8- and 72-fold increases in nirmatrelvir IC50 values. Additionally, we constructed SARS-CoV-2 omicron replicons containing these mutations, and one of the mutants (S144A/E166A) displayed a 20-fold increase in EC50, confirming the role of FEP in identifying drug-resistance mutations. Our findings suggest that FEP can be a valuable tool in proactively monitoring the emergence of resistant strains and guiding the design of future inhibitors with reduced susceptibility to drug resistance. As nirmatrelvir is currently widely used for treating COVID-19, this research has important implications for surveillance efforts and antiviral development