Artificial Intelligence. Amplify scientific discovery with artificial intelligence

Bio-Plausible Multimodal Learning with Emerging Neuromorphic Devices

Multimodal machine learning, as a prospective advancement in artificial intelligence, endeavors to emulate the brain's multimodal learning abilities with the objective to enhance interactions with humans. However, this approach requires simultaneous processing of diverse types of data, leading to increased model complexity, longer training times, and higher energy consumption. Multimodal neuromorphic devices have the capability to preprocess spatio-temporal information from various physical signals into unified electrical signals with high information density, thereby enabling more biologically plausible multimodal learning with low complexity and high energy-efficiency. Here, this work conducts a comparison between the expression of multimodal machine learning and multimodal neuromorphic computing, followed by an overview of the key characteristics associated with multimodal neuromorphic devices. The bio-plausible operational principles and the multimodal learning abilities of emerging devices are examined, which are classified into heterogeneous and homogeneous multimodal neuromorphic devices. Subsequently, this work provides a detailed description of the multimodal learning capabilities demonstrated by neuromorphic circuits and their respective applications. Finally, this work highlights the limitations and challenges of multimodal neuromorphic computing in order to hopefully provide insight into potential future research directions.

Quantifying the use and potential benefits of artificial intelligence in scientific research

The rapid advancement of artificial intelligence (AI) is poised to reshape almost every line of work. Despite enormous efforts devoted to understanding AI's economic impacts, we lack a systematic understanding of the benefits to scientific research associated with the use of AI. Here we develop a measurement framework to estimate the direct use of AI and associated benefits in science. We find that the use and benefits of AI appear widespread throughout the sciences, growing especially rapidly since 2015. However, there is a substantial gap between AI education and its application in research, highlighting a misalignment between AI expertise supply and demand. Our analysis also reveals demographic disparities, with disciplines with higher proportions of women or Black scientists reaping fewer benefits from AI, potentially exacerbating existing inequalities in science. These findings have implications for the equity and sustainability of the research enterprise, especially as the integration of AI with science continues to deepen.

Science of science: A multidisciplinary field studying science

Science and knowledge are studied by researchers across many disciplines, examining how they are developed, what their current boundaries are and how we can advance them. By integrating evidence across disparate disciplines, the holistic field of science of science can address these foundational questions. This field illustrates how science is shaped by many interconnected factors: the cognitive processes of scientists, the historical evolution of science, economic incentives, institutional influences, computational approaches, statistical, mathematical and instrumental foundations of scientific inference, scientometric measures, philosophical and ethical dimensions of scientific concepts, among other influences. Achieving a comprehensive overview of a multifaceted field like the science of science requires pulling together evidence from the many sub-fields studying science across the natural and social sciences and humanities. This enables developing an interdisciplinary perspective of scientific practice, a more holistic understanding of scientific processes and outcomes, and more nuanced perspectives to how scientific research is conducted, influenced and evolves. It enables leveraging the strengths of various disciplines to create a holistic view of the foundations of science. Different researchers study science from their own disciplinary perspective and use their own methods, and there is a large divide between quantitative and qualitative researchers as they commonly do not read or cite research using other methodological approaches. A broader, synthesizing paper employing a qualitative approach can however help provide a bridge between disciplines by pulling together aspects of science (economic, scientometric, psychological, philosophical etc.). Such an approach enables identifying, across the range of fields, the powerful role of our scientific methods and instruments in shaping most aspects of our knowledge and science, whereas economic, social and historical influences help shape what knowledge we pursue. A unifying theory is then outlined for science of science - the new-methods-drive-science theory.

Visual Knowledge Domain of Artificial Intelligence in Computed Tomography: A Review Based on Bibliometric Analysis

ABSTRACT

Artificial intelligence (AI)-assisted medical imaging technology is a new research area of great interest that has developed rapidly over the last decade. However, there has been no bibliometric analysis of published studies in this field. The present review focuses on AI-related studies on computed tomography imaging in the Web of Science database and uses CiteSpace and VOSviewer to generate a knowledge map and conduct the basic information analysis, co-word analysis, and co-citation analysis. A total of 7265 documents were included and the number of documents published had an overall upward trend. Scholars from the United States and China have made outstanding achievements, and there is a general lack of extensive cooperation in this field. In recent years, the research areas of great interest and difficulty have been the optimization and upgrading of algorithms, and the application of theoretical models to practical clinical applications. This review will help researchers understand the developments, research areas of great interest, and research frontiers in this field and provide reference and guidance for future studies.

Using machine learning to evaluate 1.2 million studies on small-scale farming and post-production food systems in low- and middle-income countries

Recent developments have emphasized the need for agrifood systems to move beyond a production-oriented approach to recognize agriculture as part of a broader agrifood system that prioritizes livelihoods, social equity, diets, and climate and environmental outcomes. At the same time, the knowledge base for agriculture is growing exponentially. Using artificial intelligence and machine learning approaches, we reviewed more than 1.2 million publications from the past 20 years to assess the current landscape of agricultural research taking place in low- and middle-income countries. The result is a clearer picture of what research has been conducted on small-scale farming and post-production systems from 2000 to the present, and where persistent evidence gaps exist. We found that the greatest focus of the literature is on economic outcomes, such as productivity, yield, and incomes. There is also some emphasis on identifying and measuring environmental outcomes. However, noticeable data gaps exist for agricultural research focused on nutrition and diet, and gender and inclusivity.

Fully nondestructive analysis of capsaicinoids electrochemistry data with deep neural network enables portable system

Electrochemical methods have been extensively applied for the detection of chemical information from food or other analytes. However, existing electrochemical methods are limited to focusing solely on the absorption peaks and disregard much of the hidden chemical fingerprint information. Consequently, electrochemical sensors are constrained by their ability to detect samples containing multiple source-material mixtures with overlapping constituents. We hypothesized that the target substances can be effectively identified and detected using differential sensor data combined with artificial intelligence (AI). In this study, we developed a novel signal array composed of five metal electrodes and used a convolutional neural network (CNN) model for feature extraction to detect capsaicinoids in stews. Our results indicate that the proposed method achieved satisfactory predictions with a root mean square error (RMSE) of 5.407 in independent brine samples. This provides a promising strategy and practical approach for the nondestructive analysis of multidimensional electrochemical data of mixed analytes.

The brain-computer analogy—“A special issue”

In this review essay, we give a detailed synopsis of the twelve contributions which are collected in a Special Issue in Frontiers Ecology and Evolution, based on the research topic “Current Thoughts on the Brain-Computer Analogy—All Metaphors Are Wrong, But Some Are Useful.” The synopsis is complemented by a graphical summary, a matrix which links articles to selected concepts. As first identified by Turing, all authors in this Special Issue recognize semantics as a crucial concern in the brain-computer analogy debate, and consequently address a number of such issues. What is missing, we believe, is the distinction between metaphor and analogy, which we reevaluate, describe in some detail, and offer a definition for the latter. To enrich the debate, we also deem necessary to develop on the evolutionary theories of the brain, of which we provide an overview. This article closes with thoughts on creativity in Science, for we concur with the stance that metaphors and analogies, and their esthetic impact, are essential to the creative process, be it in Sciences as well as in Arts.

Recent Advances in Artificial Intelligence and Tactical Autonomy: Current Status, Challenges, and Perspectives

This paper presents the findings of detailed and comprehensive technical literature aimed at identifying the current and future research challenges of tactical autonomy. It discusses in great detail the current state-of-the-art powerful artificial intelligence (AI), machine learning (ML), and robot technologies, and their potential for developing safe and robust autonomous systems in the context of future military and defense applications. Additionally, we discuss some of the technical and operational critical challenges that arise when attempting to practically build fully autonomous systems for advanced military and defense applications. Our paper provides the state-of-the-art advanced AI methods available for tactical autonomy. To the best of our knowledge, this is the first work that addresses the important current trends, strategies, critical challenges, tactical complexities, and future research directions of tactical autonomy. We believe this work will greatly interest researchers and scientists from academia and the industry working in the field of robotics and the autonomous systems community. We hope this work encourages researchers across multiple disciplines of AI to explore the broader tactical autonomy domain. We also hope that our work serves as an essential step toward designing advanced AI and ML models with practical implications for real-world military and defense settings.

Smart breeding driven by big data, artificial intelligence, and integrated genomic-enviromic prediction

The first paradigm of plant breeding involves direct selection-based phenotypic observation, followed by predictive breeding using statistical models for quantitative traits constructed based on genetic experimental design and, more recently, by incorporation of molecular marker genotypes. However, plant performance or phenotype (P) is determined by the combined effects of genotype (G), envirotype (E), and genotype by environment interaction (GEI). Phenotypes can be predicted more precisely by training a model using data collected from multiple sources, including spatiotemporal omics (genomics, phenomics, and enviromics across time and space). Integration of 3D information profiles (G-P-E), each with multidimensionality, provides predictive breeding with both tremendous opportunities and great challenges. Here, we first review innovative technologies for predictive breeding. We then evaluate multidimensional information profiles that can be integrated with a predictive breeding strategy, particularly envirotypic data, which have largely been neglected in data collection and are nearly untouched in model construction. We propose a smart breeding scheme, integrated genomic-enviromic prediction (iGEP), as an extension of genomic prediction, using integrated multiomics information, big data technology, and artificial intelligence (mainly focused on machine and deep learning). We discuss how to implement iGEP, including spatiotemporal models, environmental indices, factorial and spatiotemporal structure of plant breeding data, and cross-species prediction. A strategy is then proposed for prediction-based crop redesign at both the macro (individual, population, and species) and micro (gene, metabolism, and network) scales. Finally, we provide perspectives on translating smart breeding into genetic gain through integrative breeding platforms and open-source breeding initiatives. We call for coordinated efforts in smart breeding through iGEP, institutional partnerships, and innovative technological support.

A deep learning model identifies emphasis on hard work as an important predictor of income inequality

High levels of income inequality can persist in society only if people accept the inequality as justified. To identify psychological predictors of people's tendency to justify inequality, we retrained a pre-existing deep learning model to predict the extent to which World Values Survey respondents believed that income inequality is necessary. A feature importance analysis revealed multiple items associated with the importance of hard work as top predictors. As an emphasis on hard work is a key component of the Protestant Work Ethic, we formulated the hypothesis that the PWE increases acceptance of inequality. A correlational study found that the more people endorsed PWE, the less disturbed they were about factual statistics about wealth equality in the US. Two experiments found that exposing people to PWE items decreased their disturbance with income inequality. The findings indicate that machine learning models can be reused to generate viable hypotheses.

Autonomous materials synthesis via hierarchical active learning of nonequilibrium phase diagrams

Autonomous experimentation enabled by artificial intelligence offers a new paradigm for accelerating scientific discovery. Nonequilibrium materials synthesis is emblematic of complex, resource-intensive experimentation whose acceleration would be a watershed for materials discovery. We demonstrate accelerated exploration of metastable materials through hierarchical autonomous experimentation governed by the Scientific Autonomous Reasoning Agent (SARA). SARA integrates robotic materials synthesis using lateral gradient laser spike annealing and optical characterization along with a hierarchy of AI methods to map out processing phase diagrams. Efficient exploration of the multidimensional parameter space is achieved with nested active learning cycles built upon advanced machine learning models that incorporate the underlying physics of the experiments and end-to-end uncertainty quantification. We demonstrate SARA’s performance by autonomously mapping synthesis phase boundaries for the Bi₂O₃ system, leading to orders-of-magnitude acceleration in the establishment of a synthesis phase diagram that includes conditions for stabilizing δ-Bi₂O₃ at room temperature, a critical development for electrochemical technologies.

The application of machine learning for predicting the methane uptake and working capacity of MOFs

Multiple linear regression analysis, as a part of machine learning, is employed to develop equations for the quick and accurate prediction of the methane uptake and working capacity of metal-organic frameworks (MOFs). Only three crystal characteristics of MOFs (geometric descriptors) are employed for developing the equations: surface area, pore volume and density of the crystal structure. The values of the geometric descriptors can be obtained much more cheaply in terms of time and other resources compared to running calculations of gas sorption or performing experimental work. Within this work sets of equations are provided for the different cases studied: a series of MOFs with NbO topology, a set of benchmark MOFs with outstanding methane storage and working capacities, and the whole CoRE MOF database (11 000 structures).

An Evaluation of the Yangtze River Economic Belt Manufacturing Industry Level of Intelligentization and Influencing Factors: Evidence from China

Over recent decades, the application of artificial intelligence methods in manufacturing has led to new spheres of research such as the Internet of Things, Cyber–Physical Systems, and Cloud Computing and Big Data, leading to the so-called Industry 4.0. However, to date, little research has been geared towards assessing the factors that influence intelligent manufacturing on a regional scale. Addressing this problem, this paper constructs an evaluation index system for the Yangtze River Economic Belt (YREB) intelligent manufacturing sector using eleven years (2008–2018) of provincial panel data. The entropy method is applied to three evaluation criteria, namely intelligent innovation, equipment, and profit, to construct an evaluation index system. An analysis of the results revealed that the level intelligentization of the manufacturing industry of the YREB increases yearly, and that intelligent innovations are notably occurring at a faster rate than profits. Disproportional enterprise returns on investment have occurred, which decreases enterprise motivation to be innovative in the first place. Additionally, it was also observed that FDI, financial development, government intervention, and the level of opening-up were the primary factors modulating regional intelligent manufacturing levels.

Nobel Turing Challenge: creating the engine for scientific discovery

Scientific discovery has long been one of the central driving forces in our civilization. It uncovered the principles of the world we live in, and enabled us to invent new technologies reshaping our society, cure diseases, explore unknown new frontiers, and hopefully lead us to build a sustainable society. Accelerating the speed of scientific discovery is therefore one of the most important endeavors. This requires an in-depth understanding of not only the subject areas but also the nature of scientific discoveries themselves. In other words, the "science of science" needs to be established, and has to be implemented using artificial intelligence (AI) systems to be practically executable. At the same time, what may be implemented by "AI Scientists" may not resemble the scientific process conducted by human scientist. It may be an alternative form of science that will break the limitation of current scientific practice largely hampered by human cognitive limitation and sociological constraints. It could give rise to a human-AI hybrid form of science that shall bring systems biology and other sciences into the next stage. The Nobel Turing Challenge aims to develop a highly autonomous AI system that can perform top-level science, indistinguishable from the quality of that performed by the best human scientists, where some of the discoveries may be worthy of Nobel Prize level recognition and beyond.

Einsatz eines virtuellen Kollaborators in analogen & digitalen Workshops im organisationalen Kontext

In dieser Studie stellen wir Unterschiede und Gemeinsamkeiten in der analogen sowie digitalen Zusammenarbeit hinsichtlich eines virtuellen Kollaborators (VK) dar. Konkret beobachten wir eine Kreativeinheit in einem Industrieunternehmen sowohl in der analogen als auch in der digitalen Durchführung kollaborativer Workshops. Aus den daraus resultierenden Herausforderungen und Anforderungen, die wir anhand von Interviews erheben, leiten wir Designprinzipien an einen VK ab und ziehen einen Vergleich. Gemeinsamkeiten bestehen darin, den Teilnehmenden zusätzliche Informationen und kreativen Input aus internetbasierten Quellen zu liefern. Unterschiede bestehen in der administrativen Vor- und Nachbereitung der jeweiligen Workshops sowie in der Art der Beeinflussung kollaborativer Arbeit. Während bei der digitalen Durchführung eher die Perspektiverweiterung im Vordergrund steht, ist es bei der analogen Durchführung die Ausbalancierung der Redebeiträge. Spezifika stellen sich darüber hinaus für die digitale Durchführung bei der Vernetzung der Teilnehmenden sowie beim Umgang mit digitalen Werkzeugen.

Dielectric Polymer Property Prediction Using Recurrent Neural Networks with Optimizations

Despite the growing success of machine learning for predicting structure-property relationships in molecules and materials, such as predicting the dielectric properties of polymers, it is still in its infancy. We report on the effectiveness of solving structure-property relationships for a computer-generated database of dielectric polymers using recurrent neural network (RNN) models. The implementation of a series of optimization strategies was crucial to achieving high learning speeds and sufficient accuracy: (1) binary and nonbinary representations of SMILES (Simplified Molecular Input Line System) fingerprints and (2) backpropagation with affine transformation of the input sequence (ATransformedBP) and resilient backpropagation with initial weight update parameter optimizations (iRPROP^- optimized). For the investigated database of polymers, the binary SMILES representation was found to be superior to the decimal representation with respect to the training and prediction performance. All developed and optimized Elman-type RNN algorithms outperformed nonoptimized RNN models in the efficient prediction of nonlinear structure-activity relationships. The average relative standard deviation (RSD) remained well below 5%, and the maximum RSD did not exceed 30%. Moreover, we provide a C++ codebase as a testbed for a new generation of open programming languages that target increasingly diverse computer architectures.

The AI for Scientific Discovery Network. PATTERNS (NEW YORK, N.Y.) 2021;2:100162. [PMID: 33511363 PMCID: PMC7815949 DOI: 10.1016/j.patter.2020.100162]

The Artificial Intelligence and Augmented Intelligence for Automated Investigation for Scientific Discovery Network+ (AI3SD) was established in response to the UK Engineering and Physical Sciences Research Council (EPSRC) late-2017 call for a Network+ to promote cutting-edge research in artificial intelligence to accelerate groundbreaking scientific discoveries. This article provides the philosophical, scientific, and technical underpinnings of the Network+, the history of the different domains represented in the Network+, and the specific focus of the Network+. The activities, collaborations, and research covered in the first year of the Network+ have highlighted the significant challenges in the chemistry and augmented and artificial intelligence space. These challenges are shaping the future directions of the Network+. The article concludes with a summary of the lessons learned in running this Network+ and introduces our plans for the future in a landscape redrawn by COVID-19, including rebranding into the AI 4 Scientific Discovery Network (www.ai4science.network).

Using altmetrics for detecting impactful research in quasi-zero-day time-windows: the case of COVID-19

On December 31st 2019, the World Health Organization China Country Office was informed of cases of pneumonia of unknown etiology detected in Wuhan City. The cause of the syndrome was a new type of coronavirus isolated on January 7th 2020 and named Severe Acute Respiratory Syndrome CoronaVirus 2 (SARS-CoV-2). SARS-CoV-2 is the cause of the coronavirus disease 2019 (COVID-19). Since January 2020 an ever increasing number of scientific works related to the new pathogen have appeared in literature. Identifying relevant research outcomes at very early stages is challenging. In this work we use COVID-19 as a use-case for investigating: (1) which tools and frameworks are mostly used for early scholarly communication; (2) to what extent altmetrics can be used to identify potential impactful research in tight (i.e. quasi-zero-day) time-windows. A literature review with rigorous eligibility criteria is performed for gathering a sample composed of scientific papers about SARS-CoV-2/COVID-19 appeared in literature in the tight time-window ranging from January 15th 2020 to February 24th 2020. This sample is used for building a knowledge graph that represents the knowledge about papers and indicators formally. This knowledge graph feeds a data analysis process which is applied for experimenting with altmetrics as impact indicators. We find moderate correlation among traditional citation count, citations on social media, and mentions on news and blogs. Additionally, correlation coefficients are not inflated by indicators associated with zero values, which are quite common at very early stages after an article has been published. This suggests there is a common intended meaning of the citational acts associated with aforementioned indicators. Then, we define a method, i.e. the Comprehensive Impact Score (CIS), that harmonises different indicators for providing a multi-dimensional impact indicator. CIS shows promising results as a tool for selecting relevant papers even in a tight time-window. Our results foster the development of automated frameworks aimed at helping the scientific community in identifying relevant work even in case of limited literature and observation time.

Effective Molecular Descriptors for Chemical Accuracy at DFT Cost: Fragmentation, Error-Cancellation, and Machine Learning

Recent advances in theoretical thermochemistry have allowed the study of small organic and bio-organic molecules with high accuracy. However, applications to larger molecules are still impeded by the steep scaling problem of highly accurate quantum mechanical (QM) methods, forcing the use of approximate, more cost-effective methods at a greatly reduced accuracy. One of the most successful strategies to mitigate this error is the use of systematic error-cancellation schemes, in which highly accurate QM calculations can be performed on small portions of the molecule to construct corrections to an approximate method. Herein, we build on ideas from fragmentation and error-cancellation to introduce a new family of molecular descriptors for machine learning modeled after the Connectivity-Based Hierarchy (CBH) of generalized isodesmic reaction schemes. The best performing descriptor ML(CBH-2) is constructed from fragments preserving only the immediate connectivity of all heavy (non-H) atoms of a molecule along with overlapping regions of fragments in accordance with the inclusion-exclusion principle. Our proposed approach offers a simple, chemically intuitive grouping of atoms, tuned with an optimal amount of error-cancellation, and outperforms previous structure-based descriptors using a much smaller input vector length. For a wide variety of density functionals, DFT+ΔML(CBH-2) models, trained on a set of small- to medium-sized organic HCNOSCl-containing molecules, achieved an out-of-sample MAE within 0.5 kcal/mol and 2σ (95%) confidence interval of <1.5 kcal/mol compared to accurate G4 reference values at DFT cost.

Autonomous Discovery in the Chemical Sciences Part II: Outlook

This two-part Review examines how automation has contributed to different aspects of discovery in the chemical sciences. In this second part, we reflect on a selection of exemplary studies. It is increasingly important to articulate what the role of automation and computation has been in the scientific process and how that has or has not accelerated discovery. One can argue that even the best automated systems have yet to "discover" despite being incredibly useful as laboratory assistants. We must carefully consider how they have been and can be applied to future problems of chemical discovery in order to effectively design and interact with future autonomous platforms. The majority of this Review defines a large set of open research directions, including improving our ability to work with complex data, build empirical models, automate both physical and computational experiments for validation, select experiments, and evaluate whether we are making progress towards the ultimate goal of autonomous discovery. Addressing these practical and methodological challenges will greatly advance the extent to which autonomous systems can make meaningful discoveries.

Machine-Learning-Assisted De Novo Design of Organic Molecules and Polymers: Opportunities and Challenges

Organic molecules and polymers have a broad range of applications in biomedical, chemical, and materials science fields. Traditional design approaches for organic molecules and polymers are mainly experimentally-driven, guided by experience, intuition, and conceptual insights. Though they have been successfully applied to discover many important materials, these methods are facing significant challenges due to the tremendous demand of new materials and vast design space of organic molecules and polymers. Accelerated and inverse materials design is an ideal solution to these challenges. With advancements in high-throughput computation, artificial intelligence (especially machining learning, ML), and the growth of materials databases, ML-assisted materials design is emerging as a promising tool to flourish breakthroughs in many areas of materials science and engineering. To date, using ML-assisted approaches, the quantitative structure property/activity relation for material property prediction can be established more accurately and efficiently. In addition, materials design can be revolutionized and accelerated much faster than ever, through ML-enabled molecular generation and inverse molecular design. In this perspective, we review the recent progresses in ML-guided design of organic molecules and polymers, highlight several successful examples, and examine future opportunities in biomedical, chemical, and materials science fields. We further discuss the relevant challenges to solve in order to fully realize the potential of ML-assisted materials design for organic molecules and polymers. In particular, this study summarizes publicly available materials databases, feature representations for organic molecules, open-source tools for feature generation, methods for molecular generation, and ML models for prediction of material properties, which serve as a tutorial for researchers who have little experience with ML before and want to apply ML for various applications. Last but not least, it draws insights into the current limitations of ML-guided design of organic molecules and polymers. We anticipate that ML-assisted materials design for organic molecules and polymers will be the driving force in the near future, to meet the tremendous demand of new materials with tailored properties in different fields.

A curious formulation robot enables the discovery of a novel protocell behavior

We describe a chemical robotic assistant equipped with a curiosity algorithm (CA) that can efficiently explore the states a complex chemical system can exhibit. The CA-robot is designed to explore formulations in an open-ended way with no explicit optimization target. By applying the CA-robot to the study of self-propelling multicomponent oil-in-water protocell droplets, we are able to observe an order of magnitude more variety in droplet behaviors than possible with a random parameter search and given the same budget. We demonstrate that the CA-robot enabled the observation of a sudden and highly specific response of droplets to slight temperature changes. Six modes of self-propelled droplet motion were identified and classified using a time-temperature phase diagram and probed using a variety of techniques including NMR. This work illustrates how CAs can make better use of a limited experimental budget and significantly increase the rate of unpredictable observations, leading to new discoveries with potential applications in formulation chemistry.

Identifying high-performance catalytic conditions for carbon dioxide reduction to dimethoxymethane by multivariate modelling

In times of a warming climate due to excessive carbon dioxide production, catalytic conversion of carbon dioxide to formaldehyde is not only a process of great industrial interest, but it could also serve as a means for meeting our climate goals. Currently, formaldehyde is produced in an energetically unfavourable and atom-inefficient process. A much needed solution remains academically challenging. Here we present an algorithmic workflow to improve the ruthenium-catalysed transformation of carbon dioxide to the formaldehyde derivative dimethoxymethane. Catalytic processes are typically optimised by comprehensive screening of catalysts, substrates, reaction parameters and additives to enhance activity and selectivity. The common problem of the multidimensionality of the parameter space, leading to only incremental improvement in laborious physical investigations, was overcome by combining elements from machine learning, optimisation and experimental design, tripling the turnover number of 786 to 2761. The optimised conditions were then used in a new reaction setup tailored to the process parameters leading to a turnover number of 3874, exceeding by far those of known processes.

Comparative Analysis between International Research Hotspots and National-Level Policy Keywords on Artificial Intelligence in China from 2009 to 2018

In the last decade, artificial intelligence (AI) has undergone many important developments in China and has risen to the level of national strategy, which is closely related to the areas of research and policy promotion. The interactive relationship between the hotspots of China’s international AI research and its national-level policy keywords is the basis for further clarification and reference in academics and political circles. There has been very little research on the interaction between academic research and policy making. Understanding the relationship between the content of academic research and the content emphasized by actual operational policy will help scholars to better apply research to practice, and help decision-makers to manage effectively. Based on 3577 English publications about AI published by Chinese scholars in 2009–2018, and 262 Chinese national-level policy documents published during this period, this study carried out scientometric analysis and quantitative analysis of policy documents through the knowledge maps of AI international research hotspots in China and the co-occurrence maps of Chinese policy keywords, and conducted a comparative analysis that divided China’s AI development into three stages: the initial exploration stage, the steady rising stage, and the rapid development stage. The studies showed that in the initial exploration stage (2009–2012), research hotspots and policy keywords had a certain alienation relationship; in the steady rising stage (2013–2015), research hotspots focused more on cutting-edge technologies and policy keywords focused more on macro-guidance, and the relationship began to become close; and in the rapid development stage (2016–2018), the research hotspots and policy keywords became closely integrated, and they were mutually infiltrated and complementary, thus realizing organic integration and close connection. Through comparative analysis between international research hotspots and national-level policy keywords on AI in China from 2009 to 2018, the development of AI in China was revealed to some extent, along with the interaction between academics and politics in the past ten years, which is of great significance for the sustainable development and effective governance of China’s artificial intelligence.

Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning

Computational modeling of chemical and biological systems at atomic resolution is a crucial tool in the chemist's toolset. The use of computer simulations requires a balance between cost and accuracy: quantum-mechanical methods provide high accuracy but are computationally expensive and scale poorly to large systems, while classical force fields are cheap and scalable, but lack transferability to new systems. Machine learning can be used to achieve the best of both approaches. Here we train a general-purpose neural network potential (ANI-1ccx) that approaches CCSD(T)/CBS accuracy on benchmarks for reaction thermochemistry, isomerization, and drug-like molecular torsions. This is achieved by training a network to DFT data then using transfer learning techniques to retrain on a dataset of gold standard QM calculations (CCSD(T)/CBS) that optimally spans chemical space. The resulting potential is broadly applicable to materials science, biology, and chemistry, and billions of times faster than CCSD(T)/CBS calculations.

A White-Box Machine Learning Approach for Revealing Antibiotic Mechanisms of Action

Current machine learning techniques enable robust association of biological signals with measured phenotypes, but these approaches are incapable of identifying causal relationships. Here, we develop an integrated "white-box" biochemical screening, network modeling, and machine learning approach for revealing causal mechanisms and apply this approach to understanding antibiotic efficacy. We counter-screen diverse metabolites against bactericidal antibiotics in Escherichia coli and simulate their corresponding metabolic states using a genome-scale metabolic network model. Regression of the measured screening data on model simulations reveals that purine biosynthesis participates in antibiotic lethality, which we validate experimentally. We show that antibiotic-induced adenine limitation increases ATP demand, which elevates central carbon metabolism activity and oxygen consumption, enhancing the killing effects of antibiotics. This work demonstrates how prospective network modeling can couple with machine learning to identify complex causal mechanisms underlying drug efficacy.

Artificial intelligence outperforms experienced nephrologists to assess dry weight in pediatric patients on chronic hemodialysis

BACKGROUND

Dry weight is the lowest weight patients on hemodialysis can tolerate; correct dry weight estimation is necessary to minimize morbi-mortality, but is difficult to achieve. Here, we used artificial intelligence to improve the accuracy of dry weight assessment in hemodialysis patients.

METHODS/RESULTS

We designed a neural network which used bio-impedancemetry, blood volume monitoring, and blood pressure values as inputs; output was artificial intelligence dry weight. Fourteen pediatric patients were switched from nephrologist to artificial intelligence dry weight. Artificial intelligence dry weight was higher (28.6%), lower (50%), or identical to nephrologist dry weight. Mean difference between artificial intelligence and nephrologist dry weights was 0.497 kg (- 1.33 to + 1.29 kg). In patients for whom artificial intelligence dry weight was lower than nephrologist dry weight, systolic blood pressure significantly decreased after dry weight decrease to artificial intelligence dry weight (77th to 60th percentile, p = 0.022); anti-hypertensive treatments were successfully decreased or discontinued in 28.7% of cases. In patients for whom artificial intelligence dry weight was higher than nephrologist dry weight, no hypertension was observed after dry weight increase to artificial intelligence dry weight; when present, symptoms of dry weight underestimation receded.

CONCLUSIONS

Neural network predictions outperformed those of experienced nephrologists in most cases, proving artificial intelligence is a powerful tool for predicting dry weight in hemodialysis patients.

Controlling an organic synthesis robot with machine learning to search for new reactivity

The discovery of chemical reactions is an inherently unpredictable and time-consuming process¹. An attractive alternative is to predict reactivity, although relevant approaches, such as computer-aided reaction design, are still in their infancy². Reaction prediction based on high-level quantum chemical methods is complex³, even for simple molecules. Although machine learning is powerful for data analysis^4,5, its applications in chemistry are still being developed⁶. Inspired by strategies based on chemists' intuition⁷, we propose that a reaction system controlled by a machine learning algorithm may be able to explore the space of chemical reactions quickly, especially if trained by an expert⁸. Here we present an organic synthesis robot that can perform chemical reactions and analysis faster than they can be performed manually, as well as predict the reactivity of possible reagent combinations after conducting a small number of experiments, thus effectively navigating chemical reaction space. By using machine learning for decision making, enabled by binary encoding of the chemical inputs, the reactions can be assessed in real time using nuclear magnetic resonance and infrared spectroscopy. The machine learning system was able to predict the reactivity of about 1,000 reaction combinations with accuracy greater than 80 per cent after considering the outcomes of slightly over 10 per cent of the dataset. This approach was also used to calculate the reactivity of published datasets. Further, by using real-time data from our robot, these predictions were followed up manually by a chemist, leading to the discovery of four reactions.

Deep reinforcement learning for de novo drug design

We have devised and implemented a novel computational strategy for de novo design of molecules with desired properties termed ReLeaSE (Reinforcement Learning for Structural Evolution). On the basis of deep and reinforcement learning (RL) approaches, ReLeaSE integrates two deep neural networks-generative and predictive-that are trained separately but are used jointly to generate novel targeted chemical libraries. ReLeaSE uses simple representation of molecules by their simplified molecular-input line-entry system (SMILES) strings only. Generative models are trained with a stack-augmented memory network to produce chemically feasible SMILES strings, and predictive models are derived to forecast the desired properties of the de novo-generated compounds. In the first phase of the method, generative and predictive models are trained separately with a supervised learning algorithm. In the second phase, both models are trained jointly with the RL approach to bias the generation of new chemical structures toward those with the desired physical and/or biological properties. In the proof-of-concept study, we have used the ReLeaSE method to design chemical libraries with a bias toward structural complexity or toward compounds with maximal, minimal, or specific range of physical properties, such as melting point or hydrophobicity, or toward compounds with inhibitory activity against Janus protein kinase 2. The approach proposed herein can find a general use for generating targeted chemical libraries of novel compounds optimized for either a single desired property or multiple properties.

Objectivistic knowledge artifacts

Purpose

By establishing a conceptual path through the field of artificial intelligence for objectivistic knowledge artifacts (KAs), the purpose of this paper is to propose an extension to their design principles. The author uses these principles to deploy KAs for knowledge acquired in scientific processes, to determine whether these principles steer the design of KAs that are amenable for both human and computational manipulation.

Design/methodology/approach

Adopting the design principles mentioned above, the author describes the deployment of KAs in collaboration with a group of scientists to represent knowledge gained in scientific processes. The author then analyzes the resulting usage data.

Findings

Usage data reveal that human scientists could enter scientific KAs within the proposed structure. The scientists were able to create associations among them, search and retrieve KAs, and reuse them in drafts of reports to funding agencies. These results were observed when scientists were motivated by imminent incentives.

Research limitations/implications

Previous work has shown that objectivistic KAs are suitable for representing knowledge in computational processes. The data analyzed in this work show that they are suitable for representing knowledge in processes conducted by humans. The need for imminent incentives to motivate humans to contribute KAs suggests a limitation, which may be attributed to the exclusively objectivistic perspective in their design. The author hence discusses the adoption of situativity principles for a more beneficial implementation of KAs.

Originality/value

The suitability for interaction with both human and computational processes makes objectivistic KAs candidates for use as metadata to intersect humans and computers, particularly for scientific processes. The author found no previous work implementing objectivistic KAs for scientific knowledge.

Phenoscape: Identifying Candidate Genes for Evolutionary Phenotypes

Phenotypes resulting from mutations in genetic model organisms can help reveal candidate genes for evolutionarily important phenotypic changes in related taxa. Although testing candidate gene hypotheses experimentally in nonmodel organisms is typically difficult, ontology-driven information systems can help generate testable hypotheses about developmental processes in experimentally tractable organisms. Here, we tested candidate gene hypotheses suggested by expert use of the Phenoscape Knowledgebase, specifically looking for genes that are candidates responsible for evolutionarily interesting phenotypes in the ostariophysan fishes that bear resemblance to mutant phenotypes in zebrafish. For this, we searched ZFIN for genetic perturbations that result in either loss of basihyal element or loss of scales phenotypes, because these are the ancestral phenotypes observed in catfishes (Siluriformes). We tested the identified candidate genes by examining their endogenous expression patterns in the channel catfish, Ictalurus punctatus. The experimental results were consistent with the hypotheses that these features evolved through disruption in developmental pathways at, or upstream of, brpf1 and eda/edar for the ancestral losses of basihyal element and scales, respectively. These results demonstrate that ontological annotations of the phenotypic effects of genetic alterations in model organisms, when aggregated within a knowledgebase, can be used effectively to generate testable, and useful, hypotheses about evolutionary changes in morphology.

Inferring regulatory networks from experimental morphological phenotypes: a computational method reverse-engineers planarian regeneration

Transformative applications in biomedicine require the discovery of complex regulatory networks that explain the development and regeneration of anatomical structures, and reveal what external signals will trigger desired changes of large-scale pattern. Despite recent advances in bioinformatics, extracting mechanistic pathway models from experimental morphological data is a key open challenge that has resisted automation. The fundamental difficulty of manually predicting emergent behavior of even simple networks has limited the models invented by human scientists to pathway diagrams that show necessary subunit interactions but do not reveal the dynamics that are sufficient for complex, self-regulating pattern to emerge. To finally bridge the gap between high-resolution genetic data and the ability to understand and control patterning, it is critical to develop computational tools to efficiently extract regulatory pathways from the resultant experimental shape phenotypes. For example, planarian regeneration has been studied for over a century, but despite increasing insight into the pathways that control its stem cells, no constructive, mechanistic model has yet been found by human scientists that explains more than one or two key features of its remarkable ability to regenerate its correct anatomical pattern after drastic perturbations. We present a method to infer the molecular products, topology, and spatial and temporal non-linear dynamics of regulatory networks recapitulating in silico the rich dataset of morphological phenotypes resulting from genetic, surgical, and pharmacological experiments. We demonstrated our approach by inferring complete regulatory networks explaining the outcomes of the main functional regeneration experiments in the planarian literature; By analyzing all the datasets together, our system inferred the first systems-biology comprehensive dynamical model explaining patterning in planarian regeneration. This method provides an automated, highly generalizable framework for identifying the underlying control mechanisms responsible for the dynamic regulation of growth and form.

Developmental and regenerative biology experiments are producing a huge number of morphological phenotypes from functional perturbation experiments. However, existing pathway models do not generally explain the dynamic regulation of anatomical shape due to the difficulty of inferring and testing non-linear regulatory networks responsible for appropriate form, shape, and pattern. We present a method that automates the discovery and testing of regulatory networks explaining morphological outcomes directly from the resultant phenotypes, producing network models as testable hypotheses explaining regeneration data. Our system integrates a formalization of the published results in planarian regeneration, an in silico simulator in which the patterning properties of regulatory networks can be quantitatively tested in a regeneration assay, and a machine learning module that evolves networks whose behavior in this assay optimally matches the database of planarian results. We applied our method to explain the key experiments in planarian regeneration, and discovered the first comprehensive model of anterior-posterior patterning in planaria under surgical, pharmacological, and genetic manipulations. Beyond the planarian data, our approach is readily generalizable to facilitate the discovery of testable regulatory networks in developmental biology and biomedicine, and represents the first developmental model discovered de novo from morphological outcomes by an automated system.

Visual analysis of biological data-knowledge networks

BACKGROUND

The interpretation of the results from genome-scale experiments is a challenging and important problem in contemporary biomedical research. Biological networks that integrate experimental results with existing knowledge from biomedical databases and published literature can provide a rich resource and powerful basis for hypothesizing about mechanistic explanations for observed gene-phenotype relationships. However, the size and density of such networks often impede their efficient exploration and understanding.

RESULTS

We introduce a visual analytics approach that integrates interactive filtering of dense networks based on degree-of-interest functions with attribute-based layouts of the resulting subnetworks. The comparison of multiple subnetworks representing different analysis facets is facilitated through an interactive super-network that integrates brushing-and-linking techniques for highlighting components across networks. An implementation is freely available as a Cytoscape app.

CONCLUSIONS

We demonstrate the utility of our approach through two case studies using a dataset that combines clinical data with high-throughput data for studying the effect of β-blocker treatment on heart failure patients. Furthermore, we discuss our team-based iterative design and development process as well as the limitations and generalizability of our approach.