Pretty Darn Good Control: When are Approximate Solutions Better than Approximate Models.

Existing methods for optimal control struggle to deal with the complexity commonly encountered in real-world systems, including dimensionality, process error, model bias and data heterogeneity. Instead of tackling these system complexities directly, researchers have typically sought to simplify models to fit optimal control methods. But when is the optimal solution to an approximate, stylized model better than an approximate solution to a more accurate model? While this question has largely gone unanswered owing to the difficulty of finding even approximate solutions for complex models, recent algorithmic and computational advances in deep reinforcement learning (DRL) might finally allow us to address these questions. DRL methods have to date been applied primarily in the context of games or robotic mechanics, which operate under precisely known rules. Here, we demonstrate the ability for DRL algorithms using deep neural networks to successfully approximate solutions (the "policy function" or control rule) in a non-linear three-variable model for a fishery without knowing or ever attempting to infer a model for the process itself. We find that the reinforcement learning agent discovers a policy that outperforms both constant escapement and constant mortality policies-the standard family of policies considered in fishery management. This DRL policy has the shape of a constant escapement policy whose escapement values depend on the stock sizes of other species in the model.

Retrospective analysis of the potential use of virtual control groups in preclinical toxicity assessment using the eTOX database.

Virtual Control Groups (VCGs) based on Historical Control Data (HCD) in preclinical toxicity testing have the potential to reduce animal usage. As a case study we retrospectively analyzed the impact of replacing Concurrent Control Groups (CCGs) with VCGs on the treatment-relatedness of 28 selected histopathological findings reported in either rat or dog in the eTOX database. We developed a novel methodology whereby statistical predictions of treatment-relatedness using either CCGs or VCGs of varying covariate similarity to CCGs were compared to designations from original toxicologist reports; and changes in agreement were used to quantify changes in study outcomes. Generally, the best agreement was achieved when CCGs were replaced with VCGs with the highest level of similarity; the same species, strain, sex, administration route, and vehicle. For example, balanced accuracies for rat findings were 0.704 (predictions based on CCGs) vs. 0.702 (predictions based on VCGs). Moreover, we identified covariates which resulted in poorer identification of treatment-relatedness. This was related to an increasing incidence rate divergence in HCD relative to CCGs. Future databases which collect data at the individual animal level including study details such as animal age and testing facility are required to build adequate VCGs to accurately identify treatment-related effects.

Statistical analysis of preclinical inter-species concordance of histopathological findings in the eTOX database.

Preclinical inter-species concordance can increase the predictivity of observations to the clinic, potentially reducing drug attrition caused by unforeseen adverse events. We quantified inter-species concordance of histopathological findings and target organ toxicities across four preclinical species in the eTOX database using likelihood ratios (LRs). This was done whilst only comparing findings between studies with similar compound exposure (Δ|Cmax| ≤ 1 log-unit), repeat-dosing duration, and animals of the same sex. We discovered 24 previously unreported significant inter-species associations between histopathological findings encoded by the HPATH ontology. More associations with strong positive concordance (33% LR+ > 10) relative to strong negative concordance (12.5% LR- < 0.1) were identified. Of the top 10 most positively concordant associations, 60% were computed between different histopathological findings indicating potential differences in inter-species pathogenesis. We also observed low inter-species target organ toxicity concordance. For example, liver toxicity concordance in short-term studies between female rats and dogs observed an average LR+ of 1.84, and an average LR- of 0.73. This was corroborated by similarly low concordance between rodents and non-rodents for 75 candidate drugs in AstraZeneca. This work provides new statistically significant associations between preclinical species, but finds that concordance is rare, particularly between the absence of findings.

The economic tradeoffs and ecological impacts associated with a potential mesopelagic fishery in the California Current.

The ocean's mesopelagic zone (200-1000 m) remains one of the most understudied parts of the ocean despite knowledge that mesopelagic fishes are highly abundant. Apex predators from the surface waters are known to consume these fishes, constituting an important ecological interaction. Some countries have begun exploring the potential harvest of mesopelagic fishes to supply fishmeal and fish oil markets due to the high fish abundance in the mesopelagic zone compared with overfished surface waters. This study explored the economic and ecological implications of a moratorium on the harvest of mesopelagic fishes such as lanternfish off the US West Coast, one of the few areas where such resources are managed. We adapted a bioeconomic decision model to examine the tradeoffs between the values gained from a hypothetical mesopelagic fishery with the potential values lost from declines in predators of mesopelagic fishes facing a reduced prey resource. The economic rationale for a moratorium on harvesting mesopelagics was sensitive both to ecological relationships and the scale of the nonmarket values attributed to noncommercial predators. Using a California Current-based ecological simulation model, we found that most modeled predators of mesopelagic fishes increased in biomass even under high mesopelagic harvest rates, but the changes (either increases or decreases) were small, with relatively few predators responding with more than a 10% change in their biomass. While the ecological simulations implied that a commercial mesopelagic fishery might not have large biomass impacts for many species in the California Current system, there is still a need to further explore the various roles of the mesopelagic zone in the ocean.

Land-based measures to mitigate climate change: Potential and feasibility by country.

Land-based climate mitigation measures have gained significant attention and importance in public and private sector climate policies. Building on previous studies, we refine and update the mitigation potentials for 20 land-based measures in >200 countries and five regions, comparing "bottom-up" sectoral estimates with integrated assessment models (IAMs). We also assess implementation feasibility at the country level. Cost-effective (available up to $100/tCO2 eq) land-based mitigation is 8-13.8 GtCO2 eq yr-1 between 2020 and 2050, with the bottom end of this range representing the IAM median and the upper end representing the sectoral estimate. The cost-effective sectoral estimate is about 40% of available technical potential and is in line with achieving a 1.5°C pathway in 2050. Compared to technical potentials, cost-effective estimates represent a more realistic and actionable target for policy. The cost-effective potential is approximately 50% from forests and other ecosystems, 35% from agriculture, and 15% from demand-side measures. The potential varies sixfold across the five regions assessed (0.75-4.8 GtCO2eq yr-1 ) and the top 15 countries account for about 60% of the global potential. Protection of forests and other ecosystems and demand-side measures present particularly high mitigation efficiency, high provision of co-benefits, and relatively lower costs. The feasibility assessment suggests that governance, economic investment, and socio-cultural conditions influence the likelihood that land-based mitigation potentials are realized. A substantial portion of potential (80%) is in developing countries and LDCs, where feasibility barriers are of greatest concern. Assisting countries to overcome barriers may result in significant quantities of near-term, low-cost mitigation while locally achieving important climate adaptation and development benefits. Opportunities among countries vary widely depending on types of land-based measures available, their potential co-benefits and risks, and their feasibility. Enhanced investments and country-specific plans that accommodate this complexity are urgently needed to realize the large global potential from improved land stewardship.

Large climate mitigation potential from adding trees to agricultural lands.

While improved management of agricultural landscapes is promoted as a promising natural climate solution, available estimates of the mitigation potential are based on coarse assessments of both agricultural extent and aboveground carbon density. Here we combine 30 meter resolution global maps of aboveground woody carbon, tree cover, and cropland extent, as well as a 1 km resolution map of global pasture land, to estimate the current and potential carbon storage of trees in nonforested portions of agricultural lands. We find that global croplands currently store 3.07 Pg of carbon (C) in aboveground woody biomass (i.e., trees) and pasture lands account for an additional 3.86 Pg C across a combined 3.76 billion ha. We then estimate the climate mitigation potential of multiple scenarios of integration and avoided loss of trees in crop and pasture lands based on region-specific biomass distributions. We evaluate our findings in the context of nationally determined contributions and find that the majority of potential carbon storage from integration and avoided loss of trees in crop and pasture lands is in countries that do not identify agroforestry as a climate mitigation technique.

Does group membership affect children's judgments of social transgressions?

Children judge in-group members more favorably than out-group members. They also judge moral transgressions as more serious and more worthy of punishment than conventional transgressions. Here we asked whether children's judgments of moral and conventional transgressions vary by the group membership of the transgressor (in-group, neutral, out-group, or self). In addition, we asked whether judgments of the transgressions would extend to the transgressors themselves, including cases in which the self was the transgressor. Results show that transgressions committed by out-group members were judged as being more serious and more punish-worthy than those committed by members of other groups. In addition, children judged out-group transgressors more harshly, and the self more leniently, than other group members. Overall, results suggest that group membership does affect judgments of transgressions and transgressors, with out-group members consistently judged the most negatively and the self consistently judged the least negatively. However, when judging the transgressor or the seriousness of the transgression, domain distinctions do persist even as group membership varies. Although, when assigning punishment, domain distinctions persist only when judging the transgressions of out-group members and neutral individuals. These findings demonstrate the powerful effect of group membership on the judgments of both acts and actors, indicating that when judging transgressions children consider not only the moral or conventional status of the act but also the group membership of the actor.

Opportunities for use of one species for longer-term toxicology testing during drug development: A cross-industry evaluation.

An international expert working group representing 37 organisations (pharmaceutical/biotechnology companies, contract research organisations, academic institutions and regulatory bodies) collaborated in a data sharing exercise to evaluate the utility of two species within regulatory general toxicology studies. Anonymised data on 172 drug candidates (92 small molecules, 46 monoclonal antibodies, 15 recombinant proteins, 13 synthetic peptides and 6 antibody-drug conjugates) were submitted by 18 organisations. The use of one or two species across molecule types, the frequency for reduction to a single species within the package of general toxicology studies, and a comparison of target organ toxicities identified in each species in both short and longer-term studies were determined. Reduction to a single species for longer-term toxicity studies, as used for the development of biologicals (ICHS6(R1) guideline) was only applied for 8/133 drug candidates, but might have been possible for more, regardless of drug modality, as similar target organ toxicity profiles were identified in the short-term studies. However, definition and harmonisation around the criteria for similarity of toxicity profiles is needed to enable wider consideration of these principles. Analysis of a more robust dataset would be required to provide clear, evidence-based recommendations for expansion of these principles to small molecules or other modalities where two species toxicity testing is currently recommended.

Histopathological nerve and skeletal muscle changes in rats subjected to persistent insulin-induced hypoglycemia.

New insulin analogues with a longer duration of action and a flatter pharmacodynamic profile are developed to improve convenience and safety for diabetic patients. During the nonclinical development of such analogues, safety studies must be conducted in nondiabetic rats, which consequently are rendered chronically hypoglycemic. A rat comparator model using human insulin would be valuable, as it would enable differentiation between effects related to either persistent insulin-induced hypoglycemia (IIH) or a new analogue per se. Such a model could alleviate the need for an in-study-comparator and thereby reduce the number of animals used during development. Thus, the aims of the present study were i) to develop a preclinical animal model of persistent hypoglycemia in rats using human insulin infusion for four weeks and ii) to investigate histopathological changes in sciatic nerves and quadriceps femoris muscle tissue, as little is known about the response to persistent hypoglycemia in these tissues. Histopathologic changes in insulin-infused animals included axonal degeneration and myofibre degeneration. To our knowledge, this is the first study to show that persistent IIH provokes peripheral nerve and skeletal myofiber degeneration within the same animals. This suggests that the model can serve as a nonclinical comparator model during development of long-acting insulin analogues.

Maximizing tree carbon in croplands and grazing lands while sustaining yields.

BACKGROUND: Integrating trees into agricultural landscapes can provide climate mitigation and improves soil fertility, biodiversity habitat, water quality, water flow, and human health, but these benefits must be achieved without reducing agriculture yields. Prior estimates of carbon dioxide (CO2) removal potential from increasing tree cover in agriculture assumed a moderate level of woody biomass can be integrated without reducing agricultural production. Instead, we used a Delphi expert elicitation to estimate maximum tree covers for 53 regional cropping and grazing system categories while safeguarding agricultural yields. Comparing these values to baselines and applying spatially explicit tree carbon accumulation rates, we develop global maps of the additional CO2 removal potential of Tree Cover in Agriculture. We present here the first global spatially explicit datasets calibrated to regional grazing and croplands, estimating opportunities to increase tree cover without reducing yields, therefore avoiding a major cost barrier to restoration: the opportunity cost of CO2 removal at the expense of agriculture yields. RESULTS: The global estimated maximum technical CO2 removal potential is split between croplands (1.86 PgCO2 yr- 1) and grazing lands (1.45 PgCO2 yr- 1), with large variances. Tropical/subtropical biomes account for 54% of cropland (2.82 MgCO2 ha- 1 yr- 1, SD = 0.45) and 73% of grazing land potential (1.54 MgCO2 ha- 1 yr- 1, SD = 0.47). Potentials seem to be driven by two characteristics: the opportunity for increase in tree cover and bioclimatic factors affecting CO2 removal rates. CONCLUSIONS: We find that increasing tree cover in 2.6 billion hectares of agricultural landscapes may remove up to 3.3 billion tons of CO2 per year - more than the global annual emissions from cars. These Natural Climate Solutions could achieve the Bonn Challenge and add 793 million trees to agricultural landscapes. This is significant for global climate mitigation efforts because it represents a large, relatively inexpensive, additional CO2 removal opportunity that works within agricultural landscapes and has low economic and social barriers to rapid global scaling. There is an urgent need for policy and incentive systems to encourage the adoption of these practices.

Listening to animal behavior to understand changing ecosystems.

Interpreting sound gives powerful insight into the health of ecosystems. Beyond detecting the presence of wildlife, bioacoustic signals can reveal their behavior. However, behavioral bioacoustic information is underused because identifying the function and context of animals' sounds remains challenging. A growing acoustic toolbox is allowing researchers to begin decoding bioacoustic signals by linking individual and population-level sensing. Yet, studies integrating acoustic tools for behavioral insight across levels of biological organization remain scarce. We aim to catalyze the emerging field of behavioral bioacoustics by synthesizing recent successes and rising analytical, logistical, and ethical challenges. Because behavior typically represents animals' first response to environmental change, we posit that behavioral bioacoustics will provide theoretical and applied insights into animals' adaptations to global change.

An assessment of the state of conservation planning in Europe.

Expanding and managing current habitat and species protection measures is at the heart of the European biodiversity strategy. A structured approach is needed to gain insights into such issues is systematic conservation planning, which uses techniques from decision theory to identify places and actions that contribute most effectively to policy objectives given a set of constraints. Yet culturally and historically determined European landscapes make the implementation of any conservation plans challenging, requiring an analysis of synergies and trade-offs before implementation. In this work, we review the scientific literature for evidence of previous conservation planning approaches, highlighting recent advances and success stories. We find that the conceptual characteristics of European conservation planning studies likely reduced their potential in contributing to better-informed decisions. We outline pathways towards improving the uptake of decision theory and multi-criteria conservation planning at various scales, particularly highlighting the need for (a) open data and intuitive tools, (b) the integration of biodiversity-focused conservation planning with multiple objectives, (c) accounting of dynamic ecological processes and functions, and (d) better facilitation of entry-points and co-design practices of conservation planning scenarios with stakeholders. By adopting and improving these practices, European conservation planning might become more actionable and adaptable towards implementable policy outcomes. This article is part of the theme issue 'Ecological novelty and planetary stewardship: biodiversity dynamics in a transforming biosphere'.

Biodiversity monitoring for a just planetary future.

Data that influence policy and major investment decisions risk entrenching social and political inequities.

Historical redlining is associated with increasing geographical disparities in bird biodiversity sampling in the United States.

Historic segregation and inequality are critical to understanding modern environmental conditions. Race-based zoning policies, such as redlining in the United States during the 1930s, are associated with racial inequity and adverse multigenerational socioeconomic levels in income and education, and disparate environmental characteristics including tree canopy cover across urban neighbourhoods. Here we quantify the association between redlining and bird biodiversity sampling density and completeness-two critical metrics of biodiversity knowledge-across 195 cities in the United States. We show that historically redlined neighbourhoods remain the most undersampled urban areas for bird biodiversity today, potentially impacting conservation priorities and propagating urban environmental inequities. The disparity in sampling across redlined neighbourhood grades increased by 35.6% over the past 20 years. We identify specific urban areas in need of increased bird biodiversity sampling and discuss possible strategies for reducing uncertainty and increasing equity of sampling of biodiversity in urban areas. Our findings highlight how human behaviour and past social, economic and political conditions not just segregate our built environment but may also leave a lasting mark on the digital information we have about urban biodiversity.

Bridging adaptive management and reinforcement learning for more robust decisions.

From out-competing grandmasters in chess to informing high-stakes healthcare decisions, emerging methods from artificial intelligence are increasingly capable of making complex and strategic decisions in diverse, high-dimensional and uncertain situations. But can these methods help us devise robust strategies for managing environmental systems under great uncertainty? Here we explore how reinforcement learning (RL), a subfield of artificial intelligence, approaches decision problems through a lens similar to adaptive environmental management: learning through experience to gradually improve decisions with updated knowledge. We review where RL holds promise for improving evidence-informed adaptive management decisions even when classical optimization methods are intractable and discuss technical and social issues that arise when applying RL to adaptive management problems in the environmental domain. Our synthesis suggests that environmental management and computer science can learn from one another about the practices, promises and perils of experience-based decision-making. This article is part of the theme issue 'Detecting and attributing the causes of biodiversity change: needs, gaps and solutions'.

Differences and Similarities Between Consumer- and Caregiver- or Family-Informed Peer Roles in Mental Health.

OBJECTIVE: Peer workers are widely employed across the mental health sector in Australia, and these positions increasingly include people with experience as a service user (consumer peer workers) and people with experience as a family member (caregiver peer workers). The authors explored similarities and differences between the consumer and caregiver peer workforces and considered positions designed to combine consumer and caregiver perspectives. METHODS: A mixed-methods design was used to analyze data from a nationwide Australian survey (N=882), including responses from peer staff with consumer and caregiver perspectives (N=558), and from mental health staff not designated as peer workers (N=324). RESULTS: Most participants viewed the two perspectives as different in terms of values or goals (51%) and work practices (59%), with fewer stating that the two workforces had similar goals or values (45%) and work practices (37%). Qualitative findings provided insight into these differences and similarities, identifying differences in perspectives, priorities, and work practices but highlighting similarities in values between the two workforces. Qualitative data also revealed potential risks of employing peer workers in roles designed to use both kinds of experience for direct support roles but indicated potential for the combined perspective in other contexts. Both qualitative and quantitative data indicated that participants with consumer experience perceived greater differences between the role types than those with caregiver experience only. CONCLUSIONS: The findings indicate similarities and differences between staff with consumer or caregiver perspectives and highlight the need for greater role clarity and the potential for conflict in positions where peer workers combine both perspectives.

Direct targeting of FOXP3 in Tregs with AZD8701, a novel antisense oligonucleotide to relieve immunosuppression in cancer.

BACKGROUND: The Regulatory T cell (Treg) lineage is defined by the transcription factor FOXP3, which controls immune-suppressive gene expression profiles. Tregs are often recruited in high frequencies to the tumor microenvironment where they can suppress antitumor immunity. We hypothesized that pharmacological inhibition of FOXP3 by systemically delivered, unformulated constrained ethyl-modified antisense oligonucleotides could modulate the activity of Tregs and augment antitumor immunity providing therapeutic benefit in cancer models and potentially in man. METHODS: We have identified murine Foxp3 antisense oligonucleotides (ASOs) and clinical candidate human FOXP3 ASO AZD8701. Pharmacology and biological effects of FOXP3 inhibitors on Treg function and antitumor immunity were tested in cultured Tregs and mouse syngeneic tumor models. Experiments were controlled by vehicle and non-targeting control ASO groups as well as by use of multiple independent FOXP3 ASOs. Statistical significance of biological effects was evaluated by one or two-way analysis of variance with multiple comparisons. RESULTS: AZD8701 demonstrated a dose-dependent knockdown of FOXP3 in primary Tregs, reduction of suppressive function and efficient target downregulation in humanized mice at clinically relevant doses. Surrogate murine FOXP3 ASO, which efficiently downregulated Foxp3 messenger RNA and protein levels in primary Tregs, reduced Treg suppressive function in immune suppression assays in vitro. FOXP3 ASO promoted more than 70% reduction in FOXP3 levels in Tregs in vitro and in vivo, strongly modulated Treg effector molecules (eg, ICOS, CTLA-4, CD25 and 4-1BB), and augmented CD8+ T cell activation and produced antitumor activity in syngeneic tumor models. The combination of FOXP3 ASOs with immune checkpoint blockade further enhanced antitumor efficacy. CONCLUSIONS: Antisense inhibitors of FOXP3 offer a promising novel cancer immunotherapy approach. AZD8701 is being developed clinically as a first-in-class FOXP3 inhibitor for the treatment of cancer currently in Ph1a/b clinical trial (NCT04504669).

National mitigation potential from natural climate solutions in the tropics.

Better land stewardship is needed to achieve the Paris Agreement's temperature goal, particularly in the tropics, where greenhouse gas emissions from the destruction of ecosystems are largest, and where the potential for additional land carbon storage is greatest. As countries enhance their nationally determined contributions (NDCs) to the Paris Agreement, confusion persists about the potential contribution of better land stewardship to meeting the Agreement's goal to hold global warming below 2°C. We assess cost-effective tropical country-level potential of natural climate solutions (NCS)-protection, improved management and restoration of ecosystems-to deliver climate mitigation linked with sustainable development goals (SDGs). We identify groups of countries with distinctive NCS portfolios, and we explore factors (governance, financial capacity) influencing the feasibility of unlocking national NCS potential. Cost-effective tropical NCS offers globally significant climate mitigation in the coming decades (6.56 Pg CO2e yr-1 at less than 100 US$ per Mg CO2e). In half of the tropical countries, cost-effective NCS could mitigate over half of national emissions. In more than a quarter of tropical countries, cost-effective NCS potential is greater than national emissions. We identify countries where, with international financing and political will, NCS can cost-effectively deliver the majority of enhanced NDCs while transforming national economies and contributing to SDGs. This article is part of the theme issue 'Climate change and ecosystems: threats, opportunities and solutions'.

Proximal Neuropathy and Associated Skeletal Muscle Changes Resembling Denervation Atrophy in Hindlimbs of Chronic Hypoglycaemic Rats.

Peripheral neuropathy is one of the most common complications of diabetic hyperglycaemia. Insulin-induced hypoglycaemia (IIH) might potentially exacerbate or contribute to neuropathy as hypoglycaemia also causes peripheral neuropathy. In rats, IIH induces neuropathy associated with skeletal muscle changes. Aims of this study were to investigate the progression and sequence of histopathologic changes caused by chronic IIH in rat peripheral nerves and skeletal muscle, and whether such changes were reversible. Chronic IIH was induced by infusion of human insulin, followed by an infusion-free recovery period in some of the animals. Sciatic, plantar nerves and thigh muscle were examined histopathologically after four or eight weeks of infusion and after the recovery period. IIH resulted in high incidence of axonal degeneration in sciatic nerves and low incidence in plantar nerves indicating proximo-distal progression of the neuropathy. The neuropathy progressed in severity (sciatic nerve) and incidence (sciatic and plantar nerve) with the duration of IIH. The myopathy consisted of groups of angular atrophic myofibres which resembled histopathologic changes classically seen after denervation of skeletal muscle, and severity of the myofibre atrophy correlated with severity of axonal degeneration in sciatic nerve. Both neuropathy and myopathy were still present after four weeks of recovery, although the neuropathy was less severe. In conclusion, the results suggest that peripheral neuropathy induced by IIH progresses proximo-distally, that severity and incidence increase with duration of the hypoglycaemia and that these changes are partially reversible within four weeks. Furthermore, IIH-induced myopathy is most likely secondary to the neuropathy.