Search for: All records

The prediction of protein 3D structure from amino acid sequence is a computational grand challenge in biophysics and plays a key role in robust protein structure prediction algorithms, from drug discovery to genome interpretation. The advent of AI models, such as AlphaFold, is revolutionizing applications that depend on robust protein structure prediction algorithms. To maximize the impact, and ease the usability, of these AI tools we introduce APACE, AlphaFold2 and advanced computing as a service, a computational framework that effectively handles this AI model and its TB-size database to conduct accelerated protein structure prediction analyses in modern supercomputing environments. We deployed APACE in the Delta and Polaris supercomputers and quantified its performance for accurate protein structure predictions using four exemplar proteins: 6AWO, 6OAN, 7MEZ, and 6D6U. Using up to 300 ensembles, distributed across 200 NVIDIA A100 GPUs, we found that APACE is up to two orders of magnitude faster than off-the-self AlphaFold2 implementations, reducing time-to-solution from weeks to minutes. This computational approach may be readily linked with robotics laboratories to automate and accelerate scientific discovery. 

This study reports on the highly simple fabrication of green carbon black (GCB) generated from scrap tires with acetic acid to improve the adsorption efficiency for water purification, which is thoroughly compared with conventional carbon black (CB) obtained from petrochemicals. Unlike traditional modification processes with strong acids or bases, the introduction of a relatively mild acid readily allowed for the effective modification of GCB to increase the uptake capability of metal ions and toxic organic dyes to serve as effective adsorbents. The morphological features and thermal decomposition patterns were examined by electron microscopy and thermogravimetric analysis (TGA). The surface functional groups were characterized by Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS). The structural information (ratio of D-defects/G band-graphitic domains) obtained by Raman spectroscopy clearly suggested the successful fabrication of GCB (ID/IG ratio of 0.74), which was distinctively different from typical CB (ID/IG ratio of 0.91). In the modified GCB, the specific surface area (SBET) gradually increased with the reduction of pore size as a function of acetic acid content (52.97 m2/g for CB, 86.64 m2/g for GCB, 102.10-119.50 m2/g for acid-treated GCB). The uptake capability of the modified GCB (312.5 mg/g) for metal ions and organic dyes was greater than that of the unmodified GCB (161.3 mg/g) and typical CB (181.8 mg/g), presumably due to the presence of adsorbed acid. Upon testing them as adsorbents in an aqueous solution, all these carbon materials followed the Langmuir isotherm over the Freundlich model. In addition, the removal rates of cationic species (>70% removal of Cu2+ and crystal violet in 30 min) were much faster and far greater than those of anionic metanil yellow (<40% removal in 3 h), given the strong electrostatic interactions. Thus, this work demonstrates the possibility of recycling waste tires in the powder form of GCB as a cost-effective and green adsorbent that can potentially substitute traditional CB, and the modification strategy provides a proof of concept for developing simple fabrication guidelines of other carbonaceous materials. 

Abstract Metal-organic frameworks (MOFs) exhibit great promise for CO₂capture. However, finding the best performing materials poses computational and experimental grand challenges in view of the vast chemical space of potential building blocks. Here, we introduce GHP-MOFassemble, a generative artificial intelligence (AI), high performance framework for the rational and accelerated design of MOFs with high CO₂adsorption capacity and synthesizable linkers. GHP-MOFassemble generates novel linkers, assembled with one of three pre-selected metal nodes (Cu paddlewheel, Zn paddlewheel, Zn tetramer) into MOFs in a primitive cubic topology. GHP-MOFassemble screens and validates AI-generated MOFs for uniqueness, synthesizability, structural validity, uses molecular dynamics simulations to study their stability and chemical consistency, and crystal graph neural networks and Grand Canonical Monte Carlo simulations to quantify their CO₂adsorption capacities. We present the top six AI-generated MOFs with CO₂capacities greater than 2m mol g⁻¹, i.e., higher than 96.9% of structures in the hypothetical MOF dataset. 

Abstract We introduce an end-to-end computational framework that allows for hyperparameter optimization using theDeepHyperlibrary, accelerated model training, and interpretable AI inference. The framework is based on state-of-the-art AI models includingCGCNN,PhysNet,SchNet,MPNN,MPNN-transformer, andTorchMD-NET. We employ these AI models along with the benchmarkQM9,hMOF, andMD17datasets to showcase how the models can predict user-specified material properties within modern computing environments. We demonstrate transferable applications in the modeling of small molecules, inorganic crystals and nanoporous metal organic frameworks with a unified, standalone framework. We have deployed and tested this framework in the ThetaGPU supercomputer at the Argonne Leadership Computing Facility, and in the Delta supercomputer at the National Center for Supercomputing Applications to provide researchers with modern tools to conduct accelerated AI-driven discovery in leadership-class computing environments. We release these digital assets as open source scientific software in GitLab, and ready-to-use Jupyter notebooks in Google Colab.