Early-life physical performance predicts the aging and death of elite athletes.

Athleticism and the mortality rates begin a lifelong trajectory of decline during early adulthood. Because of the substantial follow-up time required, however, observing any longitudinal link between early-life physical declines and late-life mortality and aging remains largely inaccessible. Here, we use longitudinal data on elite athletes to reveal how early-life athletic performance predicts late-life mortality and aging in healthy male populations. Using data on over 10,000 baseball and basketball players, we calculate age at peak athleticism and rates of decline in athletic performance to predict late-life mortality patterns. Predictive capacity of these variables persists for decades after retirement, displays large effect sizes, and is independent of birth month, cohort, body mass index, and height. Furthermore, a nonparametric cohort-matching approach suggests that these mortality rate differences are associated with differential aging rates, not just extrinsic mortality. These results highlight the capacity of athletic data to predict late-life mortality, even across periods of substantial social and medical change.

Wheat physiology predictor: predicting physiological traits in wheat from hyperspectral reflectance measurements using deep learning.

BACKGROUND: The need for rapid in-field measurement of key traits contributing to yield over many thousands of genotypes is a major roadblock in crop breeding. Recently, leaf hyperspectral reflectance data has been used to train machine learning models using partial least squares regression (PLSR) to rapidly predict genetic variation in photosynthetic and leaf traits across wheat populations, among other species. However, the application of published PLSR spectral models is limited by a fixed spectral wavelength range as input and the requirement of separate custom-built models for each trait and wavelength range. In addition, the use of reflectance spectra from the short-wave infrared region requires expensive multiple detector spectrometers. The ability to train a model that can accommodate input from different spectral ranges would potentially make such models extensible to more affordable sensors. Here we compare the accuracy of prediction of PLSR with various deep learning approaches and an ensemble model, each trained and tested using previously published data sets. RESULTS: We demonstrate that the accuracy of PLSR to predict photosynthetic and related leaf traits in wheat can be improved with deep learning-based and ensemble models without overfitting. Additionally, these models can be flexibly applied across spectral ranges without significantly compromising accuracy. CONCLUSION: The method reported provides an improved prediction of wheat leaf and photosynthetic traits from leaf hyperspectral reflectance and do not require a full range, high cost leaf spectrometer. We provide a web service for deploying these algorithms to predict physiological traits in wheat from a variety of spectral data sets, with important implications for wheat yield prediction and crop breeding.

Explainable machine learning models of major crop traits from satellite-monitored continent-wide field trial data.

Four species of grass generate half of all human-consumed calories. However, abundant biological data on species that produce our food remain largely inaccessible, imposing direct barriers to understanding crop yield and fitness traits. Here, we assemble and analyse a continent-wide database of field experiments spanning 10 years and hundreds of thousands of machine-phenotyped populations of ten major crop species. Training an ensemble of machine learning models, using thousands of variables capturing weather, ground sensor, soil, chemical and fertilizer dosage, management and satellite data, produces robust cross-continent yield models exceeding R2 = 0.8 prediction accuracy. In contrast to 'black box' analytics, detailed interrogation of these models reveals drivers of crop behaviour and complex interactions predicting yield and agronomic traits. These results demonstrate the capacity of machine learning models to interrogate large datasets, generate new and testable outputs and predict crop behaviour, highlighting the powerful role of data in the future of food.

A multiple species, continent-wide, million-phenotype agronomic plant dataset.

A critical shortage of 'big' agronomic data is placing an unnecessary constraint on the conduct of public agronomic research, imparting barriers to model development and testing. Here, we address this problem by providing a large non-relational database of agronomic trials, linked to intensive management and observational data, run under a unified experimental framework. The National Variety Trials (NVTs) represent a decade-long experimental trial network, conducted across thousands of Australian field sites using highly standardised randomised controlled designs. The NVTs contain over a million machine-measured phenotypic observations, aggregated from density-controlled populations containing hundreds of millions of plants and thousands of released plant varieties. These data are linked to hundreds of thousands of metadata observations including standardised soil tests, fertiliser and pesticide input data, crop rotation data, prior farm management practices, and in-field sensors. Finally, these data are linked to a suite of ground and remote sensing observations, arranged into interpolated daily- and ten-day aggregated time series, to capture the substantial diversity in vegetation and environmental patterns across the continent-spanning NVT network.

Equitable Expanded Carrier Screening Needs Indigenous Clinical and Population Genomic Data.

Expanded carrier screening (ECS) for recessive monogenic diseases requires prior knowledge of genomic variation, including DNA variants that cause disease. The composition of pathogenic variants differs greatly among human populations, but historically, research about monogenic diseases has focused mainly on people with European ancestry. By comparison, less is known about pathogenic DNA variants in people from other parts of the world. Consequently, inclusion of currently underrepresented Indigenous and other minority population groups in genomic research is essential to enable equitable outcomes in ECS and other areas of genomic medicine. Here, we discuss this issue in relation to the implementation of ECS in Australia, which is currently being evaluated as part of the national Government's Genomics Health Futures Mission. We argue that significant effort is required to build an evidence base and genomic reference data so that ECS can bring significant clinical benefit for many Aboriginal and/or Torres Strait Islander Australians. These efforts are essential steps to achieving the Australian Government's objectives and its commitment "to leveraging the benefits of genomics in the health system for all Australians." They require culturally safe, community-led research and community involvement embedded within national health and medical genomics programs to ensure that new knowledge is integrated into medicine and health services in ways that address the specific and articulated cultural and health needs of Indigenous people. Until this occurs, people who do not have European ancestry are at risk of being, in relative terms, further disadvantaged.

ADGRL3 (LPHN3) variants predict substance use disorder.

Genetic factors are strongly implicated in the susceptibility to develop externalizing syndromes such as attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder, conduct disorder, and substance use disorder (SUD). Variants in the ADGRL3 (LPHN3) gene predispose to ADHD and predict ADHD severity, disruptive behaviors comorbidity, long-term outcome, and response to treatment. In this study, we investigated whether variants within ADGRL3 are associated with SUD, a disorder that is frequently co-morbid with ADHD. Using family-based, case-control, and longitudinal samples from disparate regions of the world (n = 2698), recruited either for clinical, genetic epidemiological or pharmacogenomic studies of ADHD, we assembled recursive-partitioning frameworks (classification tree analyses) with clinical, demographic, and ADGRL3 genetic information to predict SUD susceptibility. Our results indicate that SUD can be efficiently and robustly predicted in ADHD participants. The genetic models used remained highly efficient in predicting SUD in a large sample of individuals with severe SUD from a psychiatric institution that were not ascertained on the basis of ADHD diagnosis, thus identifying ADGRL3 as a risk gene for SUD. Recursive-partitioning analyses revealed that rs4860437 was the predominant predictive variant. This new methodological approach offers novel insights into higher order predictive interactions and offers a unique opportunity for translational application in the clinical assessment of patients at high risk for SUD.

Errors as a primary cause of late-life mortality deceleration and plateaus.

Several organisms, including humans, display a deceleration in mortality rates at advanced ages. This mortality deceleration is sufficiently rapid to allow late-life mortality to plateau in old age in several species, causing the apparent cessation of biological ageing. Here, it is shown that late-life mortality deceleration (LLMD) and late-life plateaus are caused by common demographic errors. Age estimation and cohort blending errors introduced at rates below 1 in 10,000 are sufficient to cause LLMD and plateaus. In humans, observed error rates of birth and death registration predict the magnitude of LLMD. Correction for these sources of demographic error using a mixed linear model eliminates LLMD and late-life mortality plateaus (LLMPs) without recourse to biological or evolutionary models. These results suggest models developed to explain LLMD have been fitted to an error distribution, that ageing does not slow or stop during old age in humans, and that there is a finite limit to human longevity.

Plane inclinations: A critique of hypothesis and model choice in Barbi et al.

This study highlights how the mortality plateau in Barbi and colleagues can be generated by low-frequency, randomly distributed age-misreporting errors. Furthermore, sensitivity of the late-life mortality plateau in Barbi and colleagues to the particular age range selected for regression is illustrated. Collectively, the simulation of age-misreporting errors in late-life human mortality data and a less-specific model choice than that of Barbi and colleagues highlight a clear alternative hypothesis to explanations based on evolution, the cessation of ageing, and population heterogeneity.

The dynamic upper limit of human lifespan.

We respond to claims by Dong et al. that human lifespan is limited below 125 years. Using the log-linear increase in mortality rates with age to predict the upper limits of human survival we find, in contrast to Dong et al., that the limit to human lifespan is historically flexible and increasing. This discrepancy can be explained by Dong et al.'s use of data with variable sample sizes, age-biased rounding errors, and log(0) instead of log(1) values in linear regressions. Addressing these issues eliminates the proposed 125-year upper limit to human lifespan.

Reproductive success is predicted by social dynamics and kinship in managed animal populations.

Kin and group interactions are important determinants of reproductive success in many species. Their optimization could, therefore, potentially improve the productivity and breeding success of managed populations used for agricultural and conservation purposes. Here we demonstrate this potential using a novel approach to measure and predict the effect of kin and group dynamics on reproductive output in a well-known species, the meerkat Suricata suricatta. Variation in social dynamics predicts 30% of the individual variation in reproductive success of this species in managed populations, and accurately forecasts reproductive output at least two years into the future. Optimization of social dynamics in captive meerkat populations doubles their projected reproductive output. These results demonstrate the utility of a quantitative approach to breeding programs informed by social and kinship dynamics. They suggest that this approach has great potential for improvements in the management of social endangered and agricultural species.

A new metric of inclusive fitness predicts the human mortality profile.

Biological species have evolved characteristic patterns of age-specific mortality across their life spans. If these mortality profiles are shaped by natural selection they should reflect underlying variation in the fitness effect of mortality with age. Direct fitness models, however, do not accurately predict the mortality profiles of many species. For several species, including humans, mortality rates vary considerably before and after reproductive ages, during life-stages when no variation in direct fitness is possible. Variation in mortality rates at these ages may reflect indirect effects of natural selection acting through kin. To test this possibility we developed a new two-variable measure of inclusive fitness, which we term the extended genomic output or EGO. Using EGO, we estimate the inclusive fitness effect of mortality at different ages in a small hunter-gatherer population with a typical human mortality profile. EGO in this population predicts 90% of the variation in age-specific mortality. This result represents the first empirical measurement of inclusive fitness of a trait in any species. It shows that the pattern of human survival can largely be explained by variation in the inclusive fitness cost of mortality at different ages. More generally, our approach can be used to estimate the inclusive fitness of any trait or genotype from population data on birth dates and relatedness.