HelixADMET: a robust and endpoint extensible ADMET system incorporating self-supervised knowledge transfer.

MOTIVATION: Accurate ADMET (an abbreviation for 'absorption, distribution, metabolism, excretion and toxicity') predictions can efficiently screen out undesirable drug candidates in the early stage of drug discovery. In recent years, multiple comprehensive ADMET systems that adopt advanced machine learning models have been developed, providing services to estimate multiple endpoints. However, those ADMET systems usually suffer from weak extrapolation ability. First, due to the lack of labelled data for each endpoint, typical machine learning models perform frail for the molecules with unobserved scaffolds. Second, most systems only provide fixed built-in endpoints and cannot be customized to satisfy various research requirements. To this end, we develop a robust and endpoint extensible ADMET system, HelixADMET (H-ADMET). H-ADMET incorporates the concept of self-supervised learning to produce a robust pre-trained model. The model is then fine-tuned with a multi-task and multi-stage framework to transfer knowledge between ADMET endpoints, auxiliary tasks and self-supervised tasks. RESULTS: Our results demonstrate that H-ADMET achieves an overall improvement of 4%, compared with existing ADMET systems on comparable endpoints. Additionally, the pre-trained model provided by H-ADMET can be fine-tuned to generate new and customized ADMET endpoints, meeting various demands of drug research and development requirements. AVAILABILITY AND IMPLEMENTATION: H-ADMET is freely accessible at https://paddlehelix.baidu.com/app/drug/admet/train. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Mechanistic insight into spontaneous transition from cellular alternans to arrhythmia-A simulation study.

Cardiac electrical alternans (CEA), manifested as T-wave alternans in ECG, is a clinical biomarker for predicting cardiac arrhythmias and sudden death. However, the mechanism underlying the spontaneous transition from CEA to arrhythmias remains incompletely elucidated. In this study, multiscale rabbit ventricular models were used to study the transition and a potential role of INa in perpetuating such a transition. It was shown CEA evolved into either concordant or discordant action potential (AP) conduction alternans in a homogeneous one-dimensional tissue model, depending on tissue AP duration and conduction velocity (CV) restitution properties. Discordant alternans was able to cause conduction failure in the model, which was promoted by impaired sodium channel with either a reduced or increased channel current. In a two-dimensional homogeneous tissue model, a combined effect of rate- and curvature-dependent CV broke-up alternating wavefronts at localised points, facilitating a spontaneous transition from CEA to re-entry. Tissue inhomogeneity or anisotropy further promoted break-up of re-entry, leading to multiple wavelets. Similar observations have also been seen in human atrial cellular and tissue models. In conclusion, our results identify a mechanism by which CEA spontaneously evolves into re-entry without a requirement for premature ventricular complexes or pre-existing tissue heterogeneities, and demonstrated the important pro-arrhythmic role of impaired sodium channel activity. These findings are model-independent and have potential human relevance.

A Mathematical Model of the Mouse Atrial Myocyte With Inter-Atrial Electrophysiological Heterogeneity.

Biophysically detailed mathematical models of cardiac electrophysiology provide an alternative to experimental approaches for investigating possible ionic mechanisms underlying the genesis of electrical action potentials and their propagation through the heart. The aim of this study was to develop a biophysically detailed mathematical model of the action potentials of mouse atrial myocytes, a popular experimental model for elucidating molecular and cellular mechanisms of arrhythmogenesis. Based on experimental data from isolated mouse atrial cardiomyocytes, a set of mathematical equations for describing the biophysical properties of membrane ion channel currents, intracellular Ca2+ handling, and Ca2+-calmodulin activated protein kinase II and ß-adrenergic signaling pathways were developed. Wherever possible, membrane ion channel currents were modeled using Markov chain formalisms, allowing detailed representation of channel kinetics. The model also considered heterogeneous electrophysiological properties between the left and the right atrial cardiomyocytes. The developed model was validated by its ability to reproduce the characteristics of action potentials and Ca2+ transients, matching quantitatively to experimental data. Using the model, the functional roles of four K+ channel currents in atrial action potential were evaluated by channel block simulations, results of which were quantitatively in agreement with existent experimental data. To conclude, this newly developed model of mouse atrial cardiomyocytes provides a powerful tool for investigating possible ion channel mechanisms of atrial electrical activity at the cellular level and can be further used to investigate mechanisms underlying atrial arrhythmogenesis.

The Role of CaMKII Overexpression and Oxidation in Atrial Fibrillation-A Simulation Study.

This simulation study aims to investigate how the Calcium/calmodulin-dependent protein kinase II (CaMKII) overexpression and oxidation would influence the cardiac electrophysiological behavior and its arrhythmogenic mechanism in atria. A new-built CaMKII oxidation module and a refitted CaMKII overexpression module were integrated into a mouse atrial cell model for analyzing cardiac electrophysiological variations in action potential (AP) characteristics and intracellular Ca2+ cycling under different conditions. Simulation results showed that CaMKII overexpression significantly increased the phosphorylation level of its downstream target proteins, resulting in prolonged AP and smaller calcium transient amplitude, and impaired the Ca2+ cycling stability. These effects were exacerbated by extra reactive oxygen species, which oxidized CaMKII and led to continuous high CaMKII activation in both systolic and diastolic phases. Intracellular Ca2+ depletion and sustained delayed afterdepolarizations (DADs) were observed under co-existing CaMKII overexpression and oxidation, which could be effectively reversed by clamping the phosphorylation level of ryanodine receptor (RyR). We also found that the stability of RyR release highly depended on a delicate balance between the level of RyR phosphorylation and sarcoplasmic reticulum Ca2+ concentration, which was closely related to the genesis of DADs. We concluded that the CaMKII overexpression and oxidation have a synergistic role in increasing the activity of CaMKII, and the unstable RyR may be the key downstream target in the CaMKII arrhythmogenic mechanism. Our simulation provides detailed mechanistic insights into the arrhythmogenic effect of CaMKII overexpression and oxidation, which suggests CaMKII as a promising target in the therapy of atrial fibrillation.

Pro-arrhythmic Effects of Hydrogen Sulfide in Healthy and Ischemic Cardiac Tissues: Insight From a Simulation Study.

Hydrogen sulfide (H2S), an ambient air pollutant, has been reported to increase cardiac events in patients with cardiovascular diseases, but the underlying mechanisms remain not elucidated. This study investigated the pro-arrhythmic effects of H2S in healthy and ischemic conditions. Experimental data of H2S effects on ionic channels (including the L-type Ca2+ channel and ATP-sensitive K+ channel) were incorporated into a virtual heart model to evaluate their integral action on cardiac arrhythmogenesis. It was shown that H2S depressed cellular excitability, abbreviated action potential duration, and augmented tissue's transmural dispersion of repolarization, resulting in an increase in tissue susceptibility to initiation and maintenance of reentry. The observed effects of H2S on cardiac excitation are more remarkable in the ischemic condition than in the healthy condition. This study provides mechanistic insights into the pro-arrhythmic effects of air pollution (H2S), especially in the case with extant ischemic conditions.

Quantitative proteomics and single-nucleus transcriptomics of the sinus node elucidates the foundation of cardiac pacemaking.

The sinus node is a collection of highly specialised cells constituting the heart's pacemaker. The molecular underpinnings of its pacemaking abilities are debated. Using high-resolution mass spectrometry, we here quantify >7,000 proteins from sinus node and neighbouring atrial muscle. Abundances of 575 proteins differ between the two tissues. By performing single-nucleus RNA sequencing of sinus node biopsies, we attribute measured protein abundances to specific cell types. The data reveal significant differences in ion channels responsible for the membrane clock, but not in Ca2+ clock proteins, suggesting that the membrane clock underpins pacemaking. Consistently, incorporation of ion channel expression differences into a biophysically-detailed atrial action potential model result in pacemaking and a sinus node-like action potential. Combining our quantitative proteomics data with computational modeling, we estimate ion channel copy numbers for sinus node myocytes. Our findings provide detailed insights into the unique molecular make-up of the cardiac pacemaker.

Three-dimensional image reconstruction of distribution of Pnmt+ cell-derived cells in murine heart.

Elucidating the function of specific cell types in a highly complex multicellular system such as the heart often requires detailed anatomic reconstruction. We recently described a distinctive class of phenylethanolamine n-methyltransferase (Pnmt+) cell-derived cardiomyocytes (PdCMs), a new cardiomyocyte population with a potential endocrine role. In this dataset, a 3D reconstruction was carried out to visualise the distribution of PdCMs throughout the murine heart. Rigid registration (stiff rotation and translation) was applied to properly align the fused heart slice images based on landmarks using TrakEM2, an open source plug-in in Fiji. The registered slices were then analysed and reconstructed using MATLAB (MATLAB®. Version 8.3.0.532). The final reconstructed 3D volume was 561×866×48 pixels (corresponding to spatial resolutions of 5.8, 8.9 and 2.5 mm in the x-, y- and z-direction respectively), and visualised in Paraview. The reconstruction allows for detailed analyses of morphology, projections and cellular features of different cell types, enabling further geometrical and topological analyses. Image data can be accessed and viewed through Figshare.