When we can trust computers (and when we can't)

Artificial Intelligence Must Be Made More Scientific

The role of AI within science is growing. Here we assess its impact on research and argue that AI often lacks reproducibility, transparency, objectivity, and mechanistic understanding. To ensure AI benefits research, we need to develop forms of AI that are fully compatible with the scientific method.

Multiscale co-simulation design pattern for neuroscience applications

Integration of information across heterogeneous sources creates added scientific value. Interoperability of data, tools and models is, however, difficult to accomplish across spatial and temporal scales. Here we introduce the toolbox Parallel Co-Simulation, which enables the interoperation of simulators operating at different scales. We provide a software science co-design pattern and illustrate its functioning along a neuroscience example, in which individual regions of interest are simulated on the cellular level allowing us to study detailed mechanisms, while the remaining network is efficiently simulated on the population level. A workflow is illustrated for the use case of The Virtual Brain and NEST, in which the CA1 region of the cellular-level hippocampus of the mouse is embedded into a full brain network involving micro and macro electrode recordings. This new tool allows integrating knowledge across scales in the same simulation framework and validating them against multiscale experiments, thereby largely widening the explanatory power of computational models.

A matter of trust: Learning lessons about causality will make qAOPs credible

Toxicology in the 21st Century has seen a shift from chemical risk assessment based on traditional animal tests, identifying apical endpoints and doses that are "safe", to the prospect of Next Generation Risk Assessment based on non-animal methods. Increasingly, large and high throughput in vitro datasets are being generated and exploited to develop computational models. This is accompanied by an increased use of machine learning approaches in the model building process. A potential problem, however, is that such models, while robust and predictive, may still lack credibility from the perspective of the end-user. In this commentary, we argue that the science of causal inference and reasoning, as proposed by Judea Pearl, will facilitate the development, use and acceptance of quantitative AOP models. Our hope is that by importing established concepts of causality from outside the field of toxicology, we can be "constructively disruptive" to the current toxicological paradigm, using the "Causal Revolution" to bring about a "Toxicological Revolution" more rapidly.

When Do We Need Massive Computations to Perform Detailed COVID-19 Simulations?

The COVID-19 pandemic has infected over 250 million people worldwide and killed more than 5 million as of November 2021. Many intervention strategies are utilized (e.g., masks, social distancing, vaccinations), but officials making decisions have a limited time to act. Computer simulations can aid them by predicting future disease outcomes, but they also require significant processing power or time. It is examined whether a machine learning model can be trained on a small subset of simulation runs to inexpensively predict future disease trajectories resembling the original simulation results. Using four previously published agent-based models (ABMs) for COVID-19, a decision tree regression for each ABM is built and its predictions are compared to the corresponding ABM. Accurate machine learning meta-models are generated from ABMs without strong interventions (e.g., vaccines, lockdowns) using small amounts of simulation data: the root-mean-square error (RMSE) with 25% of the data is close to the RMSE for the full dataset (0.15 vs 0.14 in one model; 0.07 vs 0.06 in another). However, meta-models for ABMs employing strong interventions require much more training data (at least 60%) to achieve a similar accuracy. In conclusion, machine learning meta-models can be used in some scenarios to assist in faster decision-making.

Using a Digital Twin of an Electrical Stimulation Device to Monitor and Control the Electrical Stimulation of Cells in vitro

Electrical stimulation for application in tissue engineering and regenerative medicine has received increasing attention in recent years. A variety of stimulation methods, waveforms and amplitudes have been studied. However, a clear choice of optimal stimulation parameters is still not available and is complicated by ambiguous reporting standards. In order to understand underlying cellular mechanisms affected by the electrical stimulation, the knowledge of the actual prevailing field strength or current density is required. Here, we present a comprehensive digital representation, a digital twin, of a basic electrical stimulation device for the electrical stimulation of cells in vitro. The effect of electrochemical processes at the electrode surface was experimentally characterised and integrated into a numerical model of the electrical stimulation. Uncertainty quantification techniques were used to identify the influence of model uncertainties on relevant observables. Different stimulation protocols were compared and it was assessed if the information contained in the monitored stimulation pulses could be related to the stimulation model. We found that our approach permits to model and simulate the recorded rectangular waveforms such that local electric field strengths become accessible. Moreover, we could predict stimulation voltages and currents reliably. This enabled us to define a controlled stimulation setting and to identify significant temperature changes of the cell culture in the monitored voltage data. Eventually, we give an outlook on how the presented methods can be applied in more complex situations such as the stimulation of hydrogels or tissue in vivo.

eScience Infrastructures in Physical Chemistry

As the volume of data associated with scientific research has exploded over recent years, the use of digital infrastructures to support this research and the data underpinning it has increased significantly. Physical chemists have been making use of eScience infrastructures since their conception, but in the last five years their usage has increased even more. While these infrastructures have not greatly affected the chemistry itself, they have in some cases had a significant impact on how the research is undertaken. The combination of the human effort of collaboration to create open source software tools and semantic resources, the increased availability of hardware for the laboratories, and the range of data management tools available has made the life of a physical chemist significantly easier. This review considers the different aspects of eScience infrastructures and explores how they have improved the way in which we can conduct physical chemistry research. Expected final online publication date for the Annual Review of Physical Chemistry, Volume 73 is April 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Pandemic drugs at pandemic speed: infrastructure for accelerating COVID-19 drug discovery with hybrid machine learning- and physics-based simulations on high-performance computers

The race to meet the challenges of the global pandemic has served as a reminder that the existing drug discovery process is expensive, inefficient and slow. There is a major bottleneck screening the vast number of potential small molecules to shortlist lead compounds for antiviral drug development. New opportunities to accelerate drug discovery lie at the interface between machine learning methods, in this case, developed for linear accelerators, and physics-based methods. The two in silico methods, each have their own advantages and limitations which, interestingly, complement each other. Here, we present an innovative infrastructural development that combines both approaches to accelerate drug discovery. The scale of the potential resulting workflow is such that it is dependent on supercomputing to achieve extremely high throughput. We have demonstrated the viability of this workflow for the study of inhibitors for four COVID-19 target proteins and our ability to perform the required large-scale calculations to identify lead antiviral compounds through repurposing on a variety of supercomputers.

Uncertainty quantification in classical molecular dynamics

Molecular dynamics simulation is now a widespread approach for understanding complex systems on the atomistic scale. It finds applications from physics and chemistry to engineering, life and medical science. In the last decade, the approach has begun to advance from being a computer-based means of rationalizing experimental observations to producing apparently credible predictions for a number of real-world applications within industrial sectors such as advanced materials and drug discovery. However, key aspects concerning the reproducibility of the method have not kept pace with the speed of its uptake in the scientific community. Here, we present a discussion of uncertainty quantification for molecular dynamics simulation designed to endow the method with better error estimates that will enable it to be used to report actionable results. The approach adopted is a standard one in the field of uncertainty quantification, namely using ensemble methods, in which a sufficiently large number of replicas are run concurrently, from which reliable statistics can be extracted. Indeed, because molecular dynamics is intrinsically chaotic, the need to use ensemble methods is fundamental and holds regardless of the duration of the simulations performed. We discuss the approach and illustrate it in a range of applications from materials science to ligand-protein binding free energy estimation. This article is part of the theme issue 'Reliability and reproducibility in computational science: implementing verification, validation and uncertainty quantification in silico'.