RevGraphVAMP: A protein molecular simulation analysis model combining graph convolutional neural networks and physical constraints.

Molecular dynamics simulation is a crucial research domain within the life sciences, focusing on comprehending the mechanisms of biomolecular interactions at atomic scales. Protein simulation, as a critical subfield, often utilizes MD for implementation, with trajectory data play a pivotal role in drug discovery. The advancement of high-performance computing and deep learning technology becomes popular and critical to predict protein properties from vast trajectory data, posing challenges regarding data features extraction from the complicated simulation data and dimensionality reduction. Simultaneously, it is essential to provide a meaningful explanation of the biological mechanism behind dimensionality. To tackle this challenge, we propose a new unsupervised model named RevGraphVAMP to intelligently analyze the simulation trajectory. This model is based on the variational approach for Markov processes (VAMP) and integrates graph convolutional neural networks and physical constraint optimization to enhance the learning performance. Additionally, we introduce attention mechanism to assess the importance of key interaction region, facilitating the interpretation of molecular mechanism. In comparison to other VAMPNets models, our model showcases competitive performance, improved accuracy in state transition prediction, as demonstrated through its application to two public datasets and the Shank3-Rap1 complex, which is associated with autism spectrum disorder. Moreover, it enhanced dimensionality reduction discrimination across different substates and provides interpretable results for protein structural characterization.

Nonlinear Relationship Between Low Density Lipoprotein and the Probability of Diabetic Macular Edema.

Purpose: Previous studies simply linearized the relationship between low density lipoprotein (LDL) and diabetic macular edema's (DME) probability, ignoring the possibility of a nonlinear relationship between them. We aimed to investigate the nonlinear relationship between LDL and DME probability in patients with type 2 diabetes mellitus (T2DM). Patients and methods: The study recruited 431 T2DM patients who attended Guangdong Provincial People's Hospital from December 2017 to November 2018. A multivariate logistic regression model was conducted to evaluate the association between LDL and DME probability. The nonlinear relationship was identified by generalized additive model. Subgroup analyses were performed to assess the consistency of the association in different subgroups. Results: LDL was positively associated with DME probability (OR=1.60, 95% CI: 1.10~2.34, P=0.0145) after adjusting for covariates. A nonlinear relationship between LDL and DME probability was discovered, with an inflection point for LDL around 4.85 mmol/L (95% CI: 4.18~4.93, P=0.037). The effect sizes and the confidence intervals on the left and right sides of inflection point were 2.17 (1.31 to 3.58) and 0.26 (0.04 to 1.77), respectively. Subgroup analyses revealed other variables had no effect on the association between them. Conclusion: Our finding suggested LDL was positively correlated with DME probability in T2DM patients. And the relationship between LDL and DME probability was nonlinear. Our findings need to be confirmed by further causal researches.

Molecular basis for substrate transport of Mycobacterium tuberculosis ABC importer DppABCD.

The type I adenosine 5'-triphosphate (ATP)-binding cassette (ABC) transporter DppABCD is believed to be responsible for the import of exogenous heme as an iron source into the cytoplasm of the human pathogen Mycobacterium tuberculosis (Mtb). Additionally, this system is also known to be involved in the acquisition of tri- or tetra-peptides. Here, we report the cryo-electron microscopy structures of the dual-function Mtb DppABCD transporter in three forms, namely, the apo, substrate-bound, and ATP-bound states. The apo structure reveals an unexpected and previously uncharacterized assembly mode for ABC importers, where the lipoprotein DppA, a cluster C substrate-binding protein (SBP), stands upright on the translocator DppBCD primarily through its hinge region and N-lobe. These structural data, along with biochemical studies, reveal the assembly of DppABCD complex and the detailed mechanism of DppABCD-mediated transport. Together, these findings provide a molecular roadmap for understanding the transport mechanism of a cluster C SBP and its translocator.

An oligopeptide permease, OppABCD, requires an iron-sulfur cluster domain for functionality.

Oligopeptide permease, OppABCD, belongs to the type I ABC transporter family. Its role is to import oligopeptides into bacteria for nutrient uptake and to modulate the host immune response. OppABCD consists of a cluster C substrate-binding protein (SBP), OppA, membrane-spanning OppB and OppC subunits, and an ATPase, OppD, that contains two nucleotide-binding domains (NBDs). Here, using cryo-electron microscopy, we determined the high-resolution structures of Mycobacterium tuberculosis OppABCD in the resting state, oligopeptide-bound pre-translocation state, AMPPNP-bound pre-catalytic intermediate state and ATP-bound catalytic intermediate state. The structures show an assembly of a cluster C SBP with its ABC translocator and a functionally required [4Fe-4S] cluster-binding domain in OppD. Moreover, the ATP-bound OppABCD structure has an outward-occluded conformation, although no substrate was observed in the transmembrane cavity. Here, we reveal an oligopeptide recognition and translocation mechanism of OppABCD, which provides a perspective on how this and other type I ABC importers facilitate bulk substrate transfer across the lipid bilayer.

SWsnn: A Novel Simulator for Spiking Neural Networks.

Spiking neural network (SNN) simulators play an important role in neural system modeling and brain function research. They can help scientists reproduce and explore neuronal activities in brain regions, neuroscience, brain-like computing, and other fields and can also be applied to artificial intelligence, machine learning, and other fields. At present, many simulators using central processing unit (CPU) or graphics processing unit (GPU) have been developed. However, due to the randomness of connections between neurons and spiking events in SNN simulation, this causes a lot of memory access time. To alleviate this problem, we developed an SNN simulator SWsnn based on the new Sunway SW26010pro processor. The SW26010pro processor consists of six core groups, each with 16 MB of local data memory (LDM). LDM has the characteristics of high-speed read and write, which is suitable for performing simulation tasks similar to SNNs. Experimental results show that SWsnn runs faster than other mainstream GPU-based simulators when simulating a certain scale of neural network, showing a strong performance advantage. To conduct larger scale simulations, SWsnn designed a simulation computation based on a large shared model of Sunway processor and developed a multiprocessor version of SWsnn based on this mode, achieving larger scale SNN simulations.