NNP/MM: Accelerating Molecular Dynamics Simulations with Machine Learning Potentials and Molecular Mechanics.

Machine learning potentials have emerged as a means to enhance the accuracy of biomolecular simulations. However, their application is constrained by the significant computational cost arising from the vast number of parameters compared with traditional molecular mechanics. To tackle this issue, we introduce an optimized implementation of the hybrid method (NNP/MM), which combines a neural network potential (NNP) and molecular mechanics (MM). This approach models a portion of the system, such as a small molecule, using NNP while employing MM for the remaining system to boost efficiency. By conducting molecular dynamics (MD) simulations on various protein-ligand complexes and metadynamics (MTD) simulations on a ligand, we showcase the capabilities of our implementation of NNP/MM. It has enabled us to increase the simulation speed by ∼5 times and achieve a combined sampling of 1 µs for each complex, marking the longest simulations ever reported for this class of simulations.

Population-based analysis of radiotherapy and chemotherapy treatment in the last month of life within regional Australia.

BACKGROUND: Cancer treatment near end of life is not likely to add meaningful benefit and minimising intervention rates has been promoted as an indicator of quality of care. Population-based analysis of treatment allows comparative analysis of treatment rates and provides insight into patterns of care. AIMS: To report a population-based analysis of both radiotherapy and active systemic therapy (AST) delivery rates along with patterns of treatment within the last 14 and 30 days of life. METHODS: The Evaluation of Cancer Outcomes Registry records clinical information on all newly diagnosed cancer patients for the Barwon South West Region of Victoria, Australia. Diagnosis details, tumour type and stage as well as core treatment details and date of death were extracted for all patients diagnosed from 2009 to 2015 inclusive. RESULTS: A total of 12 760 cases cancers were recorded. The median age of all cases was 68.8, and 53% were male. AST was received by 3699 (29%) of patients and radiotherapy by 3811 (30%). Patient deaths within 14 and 30 days of treatment for AST were 4.3 and 8.7%, respectively, and deaths within 14 and 30 days of treatment for radiotherapy 3.8 and 8.0% respectively. Factors associated with death within 30 days of AST and/or radiotherapy were male gender, age greater than 70 years and higher disease stage (all P < 0.01). Treatment rates within 30 days of death were highest for lung cancer (23% of cases) and lowest for breast cancer (2% of cases). CONCLUSIONS: This population-based analysis of AST and radiotherapy treatment within the last 30 days of life within a region of Australia has shown overall treatment rates below 10%. Treatment rates appear influenced by both patient and tumour characteristics. Future focus on subgroups with high rates of late intervention may help minimise treatment unlikely to add benefit.

Solving the RNA design problem with reinforcement learning.

We use reinforcement learning to train an agent for computational RNA design: given a target secondary structure, design a sequence that folds to that structure in silico. Our agent uses a novel graph convolutional architecture allowing a single model to be applied to arbitrary target structures of any length. After training it on randomly generated targets, we test it on the Eterna100 benchmark and find it outperforms all previous algorithms. Analysis of its solutions shows it has successfully learned some advanced strategies identified by players of the game Eterna, allowing it to solve some very difficult structures. On the other hand, it has failed to learn other strategies, possibly because they were not required for the targets in the training set. This suggests the possibility that future improvements to the training protocol may yield further gains in performance.

Integrated respiratory and palliative care leads to high levels of satisfaction: a survey of patients and carers.

BACKGROUND: The Advanced Lung Disease Service is a unique, new model of integrated respiratory and palliative care, which aims to address the unmet needs of patients with advanced, non-malignant, respiratory diseases. This study aimed to explore patients' and carers' experiences of integrated palliative care and identify valued aspects of care. METHODS: All current patients of the integrated service and their carers were invited to complete a confidential questionnaire by post or with an independent researcher. RESULTS: Eighty-eight responses were received from 64 (80.0%) eligible patients and from 24 (60%) eligible carers. Most participants (84, 95.5%) believed the integrated service helped them to manage breathlessness and nearly all participants (87, 98.9%) reported increased confidence managing symptoms. One third of patients (34.4%) had received a nurse-led domiciliary visit, with nearly all regarding this as helpful. Most participants believed the integrated respiratory and palliative care team listened to them carefully (87, 98.9%) with opportunities to express their views (88, 100%). Highly valued aspects of the service were continuity of care (82, 93.2%) and long-term care (77, 87.5%). Three quarters of participants (66, 75.0%) rated their care as excellent, with 20.5% rating it as very good. Nearly all (87, 98.9%) participants reported that they would recommend the service to others. CONCLUSIONS: Patients and carers expressed high levels of satisfaction with this model of integrated respiratory and palliative care. Continuity of care, high quality communication and feeling cared for were greatly valued and highlight simple but important aspects of care for all patients.

Statistically Rigorous Silver Nanowire Diameter Distribution Quantification by Automated Electron Microscopy and Image Analysis.

Silver nanowire (AgNW) diameters are typically characterized by manual measurement from high magnification electron microscope images. Measurement is monotonous and has potential ergonomic hazards. Because of this, statistics regarding wire diameter distribution can be poor, costly, and low-throughput. In addition, manual measurements are of unknown uncertainty and operator bias. In this paper we report an improved microscopy method for diameter and yield measurement of nanowires in terms of speed/automation and reduction of analyst variability. Each step in the process to generate these measurements was analyzed and optimized: microscope imaging conditions, sample preparation for imaging, image acquisition, image analysis, and data processing. With the resulting method, average diameter differences between samples of just a few nanometers can be confidently and statistically distinguished, allowing the identification of subtle incremental improvements in reactor processing conditions, and insight into nucleation and growth kinetics of AgNWs.

The deteriorating patient representative on a palliative care quality committee: Ethical and practical considerations.

This case study describes the involvement of a patient representative on a palliative care committee and outlines some of the issues that arose as her health deteriorated. A summary of the increasing involvement of patient representation within health care governance is provided, and some of the challenges raised by the case, many of which may be relatively unique to palliative care, are discussed. It is hoped that presentation of this fairly novel scenario provides other palliative care providers with the opportunity to consider their own processes and practices around managing a similar situation should it occur in their healthcare setting.

Use of clinical quality indicators to improve lung cancer care in a regional/rural network of health services.

PROBLEM: Optimal lung cancer care requires multidisciplinary team input, with access to specialised diagnostic and therapeutic services that may be limited in rural or regional areas and impact clinical outcomes. Clinical quality indicators can be used to measure the quality of care delivered to patients with lung cancer in a region and identify areas for improvement. We describe the implementation of internationally recognised clinical quality indicators for lung cancer care in the Barwon South Western region. DESIGN: The consensus of an expert panel was used for the selection of clinical quality indicators. The data were retrospectively collected from the Evaluation of Cancer Outcomes Barwon South West Registry, which systematically records detailed information on all new patients with cancer in the region. SETTING: Region-based health service. KEY MEASURES FOR IMPROVEMENT: Adherence to clinical quality indicator targets. STRATEGIES FOR CHANGE: Clinical quality indicators, which fall short of the expected targets, highlight areas for improvement in the service provided to patients with lung cancer. These results have prompted changes in the service offered to these patients, such as the introduction of a multidisciplinary lung cancer clinic. EFFECTS OF CHANGE: The multidisciplinary lung cancer clinic has streamlined the access to lung cancer services, including specialist consultations, diagnostics and therapeutic services, in a regional setting. Ongoing data collection is required to determine the effect of such changes on adherence to clinical quality indicator targets. LESSONS LEARNT: The regular monitoring of clinical quality indicators serves as a useful method of quality assurance in the care of patients with lung cancer. We expect these clinical quality indicators to also be used by other health services to analyse and improve services provided to patients with lung cancer.

Molecular dynamics simulations using the drude polarizable force field on GPUs with OpenMM: Implementation, validation, and benchmarks.

Presented is the implementation of the Drude force field in the open-source OpenMM simulation package allowing for access to graphical processing unit (GPU) hardware. In the Drude model, electronic degrees of freedom are represented by negatively charged particles attached to their parent atoms via harmonic springs, such that extra computational overhead comes from these additional particles and virtual sites representing lone pairs on electronegative atoms, as well as the associated thermostat and integration algorithms. This leads to an approximately fourfold increase in computational demand over additive force fields. However, by making the Drude model accessible to consumer-grade desktop GPU hardware it will be possible to perform simulations of one microsecond or more in less than a month, indicating that the barrier to employ polarizable models has largely been removed such that polarizable simulations with the classical Drude model are readily accessible and practical.

OpenMM 7: Rapid development of high performance algorithms for molecular dynamics.

OpenMM is a molecular dynamics simulation toolkit with a unique focus on extensibility. It allows users to easily add new features, including forces with novel functional forms, new integration algorithms, and new simulation protocols. Those features automatically work on all supported hardware types (including both CPUs and GPUs) and perform well on all of them. In many cases they require minimal coding, just a mathematical description of the desired function. They also require no modification to OpenMM itself and can be distributed independently of OpenMM. This makes it an ideal tool for researchers developing new simulation methods, and also allows those new methods to be immediately available to the larger community.

Medication management for community palliative care patients and the role of a specialist palliative care pharmacist: A qualitative exploration of consumer and health care professional perspectives.

BACKGROUND: Pharmacists have a key role to play in optimisation of medication regimens and promotion of medication safety. The role of specialist pharmacists as part of the multidisciplinary palliative care team, especially in the primary care setting, is not widely recognised. AIM: To explore the perspectives of stakeholders about the gaps in the current model of community palliative care services in relation to medication management and to assess their opinions pertaining to the role of a specialist palliative care pharmacist in addressing some of those gaps. DESIGN: Qualitative study utilising three focus groups involving 20 stakeholders. Thematic analysis was carried out using a framework approach and interpreted in the context of the Chronic Care Model for improving primary care for patients with chronic illness. SETTING/PARTICIPANTS: Setting was a large regional Australian palliative care service. Participants included palliative care consumers and clinicians specifically patients, caregivers, physicians, nurses and pharmacists. RESULTS: Five major themes emerged from the focus groups: access to resources, medicines and information; shared care; challenges of polypharmacy; informal caregiver needs and potential roles of a palliative care pharmacist. Gaps in access to medicines/resources, training for generalist practitioners, communication between treating teams and lack of support for patients and carers were cited as factors adversely impacting medication management in community-based palliative care. CONCLUSION: While community-based palliative care is an essential aspect of meeting the health care demands of an ageing society, the current model has several gaps and limitations. An appropriately qualified and skilled pharmacist within the palliative care team may help to address some of the gaps in relation to medication access and appropriateness.

MSMBuilder: Statistical Models for Biomolecular Dynamics.

MSMBuilder is a software package for building statistical models of high-dimensional time-series data. It is designed with a particular focus on the analysis of atomistic simulations of biomolecular dynamics such as protein folding and conformational change. MSMBuilder is named for its ability to construct Markov state models (MSMs), a class of models that has gained favor among computational biophysicists. In addition to both well-established and newer MSM methods, the package includes complementary algorithms for understanding time-series data such as hidden Markov models and time-structure based independent component analysis. MSMBuilder boasts an easy to use command-line interface, as well as clear and consistent abstractions through its Python application programming interface. MSMBuilder was developed with careful consideration for compatibility with the broader machine learning community by following the design of scikit-learn. The package is used primarily by practitioners of molecular dynamics, but is just as applicable to other computational or experimental time-series measurements.

Efficient gaussian density formulation of volume and surface areas of macromolecules on graphical processing units.

We present an algorithm to efficiently compute accurate volumes and surface areas of macromolecules on graphical processing unit (GPU) devices using an analytic model which represents atomic volumes by continuous Gaussian densities. The volume of the molecule is expressed by means of the inclusion-exclusion formula, which is based on the summation of overlap integrals among multiple atomic densities. The surface area of the molecule is obtained by differentiation of the molecular volume with respect to atomic radii. The many-body nature of the model makes a port to GPU devices challenging. To our knowledge, this is the first reported full implementation of this model on GPU hardware. To accomplish this, we have used recursive strategies to construct the tree of overlaps and to accumulate volumes and their gradients on the tree data structures so as to minimize memory contention. The algorithm is used in the formulation of a surface area-based non-polar implicit solvent model implemented as an open source plug-in (named GaussVol) for the popular OpenMM library for molecular mechanics modeling. GaussVol is 50 to 100 times faster than our best optimized implementation for the CPUs, achieving speeds in excess of 100 ns/day with 1 fs time-step for protein-sized systems on commodity GPUs. © 2017 Wiley Periodicals, Inc.

Palliative care in general practice: GP integration in caring for patients with advanced cancer.

BACKGROUND: Patients with advanced cancer often desire home-based care, placing general practitioners (GPs) at the centre of complex clinical situations. The objective of this article was to determine GPs' needs when providing home-based palliative care in collaboration with existing palliative care services. METHODS: A survey of GPs was conducted to determine knowledge, skills and confidence in providing community-based palliative care. RESULTS: Of the 56 respondents, 82% reported that they were involved in palliative management of at least one cancer patient in the previous year. A significant number of GPs (31%) lacked confidence in providing this care because of patient complexity, inadequate training and insufficient resources. Other barriers included poor communication from specialists and treating teams. Factors facilitating provision of home-based palliative care were community palliative care services and links to hospital-based palliative care teams. DISCUSSION: This survey highlights the importance of support and resources to empower GPs to confidently provide home-based palliative care for patients with advanced cancer.

OpenMM: A Hardware Independent Framework for Molecular Simulations.

The wide diversity of computer architectures today requires a new approach to software development. OpenMM is a framework for molecular mechanics simulations, allowing a single program to run efficiently on a variety of hardware platforms.

TorchMD-Net 2.0: Fast Neural Network Potentials for Molecular Simulations.

Achieving a balance between computational speed, prediction accuracy, and universal applicability in molecular simulations has been a persistent challenge. This paper presents substantial advancements in TorchMD-Net software, a pivotal step forward in the shift from conventional force fields to neural network-based potentials. The evolution of TorchMD-Net into a more comprehensive and versatile framework is highlighted, incorporating cutting-edge architectures such as TensorNet. This transformation is achieved through a modular design approach, encouraging customized applications within the scientific community. The most notable enhancement is a significant improvement in computational efficiency, achieving a very remarkable acceleration in the computation of energy and forces for TensorNet models, with performance gains ranging from 2× to 10× over previous, nonoptimized, iterations. Other enhancements include highly optimized neighbor search algorithms that support periodic boundary conditions and smooth integration with existing molecular dynamics frameworks. Additionally, the updated version introduces the capability to integrate physical priors, further enriching its application spectrum and utility in research. The software is available at https://github.com/torchmd/torchmd-net.

TorchMD-Net 2.0: Fast Neural Network Potentials for Molecular Simulations.

Achieving a balance between computational speed, prediction accuracy, and universal applicability in molecular simulations has been a persistent challenge. This paper presents substantial advancements in the TorchMD-Net software, a pivotal step forward in the shift from conventional force fields to neural network-based potentials. The evolution of TorchMD-Net into a more comprehensive and versatile framework is highlighted, incorporating cutting-edge architectures such as TensorNet. This transformation is achieved through a modular design approach, encouraging customized applications within the scientific community. The most notable enhancement is a significant improvement in computational efficiency, achieving a very remarkable acceleration in the computation of energy and forces for Tensor-Net models, with performance gains ranging from 2x to 10x over previous, non-optimized, iterations. Other enhancements include highly optimized neighbor search algorithms that support periodic boundary conditions and smooth integration with existing molecular dynamics frameworks. Additionally, the updated version introduces the capability to integrate physical priors, further enriching its application spectrum and utility in research. The software is available at https://github.com/torchmd/torchmd-net.

OpenMM 8: Molecular Dynamics Simulation with Machine Learning Potentials.

Machine learning plays an important and growing role in molecular simulation. The newest version of the OpenMM molecular dynamics toolkit introduces new features to support the use of machine learning potentials. Arbitrary PyTorch models can be added to a simulation and used to compute forces and energy. A higher-level interface allows users to easily model their molecules of interest with general purpose, pretrained potential functions. A collection of optimized CUDA kernels and custom PyTorch operations greatly improves the speed of simulations. We demonstrate these features in simulations of cyclin-dependent kinase 8 (CDK8) and the green fluorescent protein chromophore in water. Taken together, these features make it practical to use machine learning to improve the accuracy of simulations with only a modest increase in cost.