• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
              Advocacy & Research for Unlimited Lifespans


Age and life expectancy clocks based on machine learning analysis of mouse frailty

machine learning mice aging

  • Please log in to reply
No replies to this topic

#1 Engadin

  • Guest
  • 196 posts
  • 559
  • Location:Madrid
  • NO

Posted 27 December 2019 - 07:31 PM







F U L L   T E X T   S O U R C E :   bioRXiv








The identification of genes and interventions that slow or reverse aging is hampered by the lack of non-invasive metrics that can predict life expectancy of pre-clinical models. Frailty Indices (FIs) in mice are composite measures of health that are cost-effective and non-invasive, but whether they can accurately predict health and lifespan is not known. Here, mouse FIs were scored longitudinally until death and machine learning was employed to develop two clocks. A random forest regression was trained on FI components for chronological age to generate the FRIGHT (Frailty Inferred Geriatric Health Timeline) clock, a strong predictor of chronological age. A second model was trained on remaining lifespan to generate the AFRAID (Analysis of Frailty and Death) clock, which accurately predicts life expectancy and the efficacy of a lifespan-extending intervention up to a year in advance. Adoption of these clocks should accelerate the identification of novel longevity genes and aging interventions.
Aging is a biological process that causes physical and physiological deficits over time, culminating in organ failure and death. For species that experience aging, which includes nearly all animals, its presentation is not uniform; individuals age at different rates and in different ways. Biological age is an increasingly utilized concept that aims to more accurately reflect aging in an individual than the conventional chronological age. Biological measures that accurately predict health and longevity would greatly expedite studies aimed at identifying novel genetic and pharmacological disease and aging interventions.
Any useful biometric or biomarker for biological age should track with chronological age and should serve as a better predictor of remaining longevity and other age-associated outcomes than does chronological age alone, even at an age when most of a population is still alive. In addition, its measurement should be non-invasive to allow for repeated measurements without altering the health or lifespan of the animal measured (Butler et al. 2004). In humans, biometrics and biomarkers that meet at least some of these requirements include physiological measurements such as grip strength or gait (Rantanen et al. 2000; Bittner et al. 1993), measures of the immune system (Alpert et al. 2019; Martínez de Toda et al. 2019), telomere length (Mather et al. 2011), advanced glycosylation end-products (Krištić et al. 2014), levels of cellular senescence (Wang and Dreesen 2018), and DNA methylation clocks (Horvath 2013b). DNA methylation clocks have been adapted for mice but unfortunately these clocks are currently expensive, time consuming, and require the extraction of blood or tissue.
Frailty index assessments in humans are strong predictors of mortality and morbidity, outperforming other measures of biological age including DNA methylation clocks (Kim et al. 2017; Horvath and Raj 2018). Frailty indices quantify the accumulation of up to 70 health-related deficits, including laboratory test results, symptoms, diseases and standard measures such as activities of daily living (Searle et al. 2008; Mitnitski, Mogilner, and Rockwood 2001). The number of deficits an individual shows is divided by the number of items measured to give a number between 0 and 1, in which a higher number indicates a greater degree of frailty. The frailty index has been recently reverse-translated into an assessment tool for mice which includes 31 non-invasive items across a range of systems (Whitehead et al. 2014). The mouse frailty index is strongly associated with chronological age (Whitehead et al. 2014; Kane et al. 2018), correlated with mortality and other age-related outcomes (Feridooni et al. 2017; Rockwood et al. 2017), and is sensitive to lifespan-altering interventions (Kane et al. 2016). However, the power of the mouse frailty index to model biological age or predict life expectancy for an individual animal has not yet been explored.
In this study, we tracked frailty longitudinally in a cohort of aging male mice from 21 months of age until their natural deaths and employed machine learning algorithms to build two clocks: FRIGHT age, designed to model chronological age, and the AFRAID clock, which was modelled to predict life expectancy. FRIGHT age reflects apparent chronological age better than FI alone, while the AFRAID clock predicts life expectancy at multiple ages. These clocks were then tested for their predicitve power on cohorts of mice treated with interventions known to extend healthspan or lifespan, enalapril and methionine restriction. They accurately predicted increased healthspan and lifespan, demonstrating that an assessment of non-invasive biometrics in interventional studies can greatly accelerate the pace of discovery.
Frailty correlates with and is predictive of age
We measured FI scores (Figure S1) approximately every six weeks in a population of naturally aging male C57BL/6Nia mice (n=51) until the end of their lives. These mice had a normal lifespan, with a median survival of 30 months and a maximum (90th percentile) of 36 months (Figure 1a, Figure S2). As expected, FI scores increased with age from 21 to 36 months at the population level (Figure 1b). At the individual level, frailty trajectories displayed significant variance, representative of the variability in how individuals experience aging within a population of inbred animals (Figure 1c). As FI score was well-correlated with chronological age, we sought to determine the degree to which FI score could model chronological and biological age. We performed a linear regression on FI score for age with a training dataset and evaluated its accuracy on a testing dataset (Figure 1d-e). FI score was able to predict chronological age with a median error of 1.8 months and an r-squared value of 0.642 (p=7.3e-20). We hypothesized that the error may be representative of biological age, with healthier individuals having a predicted age younger than their true age. We calculated this difference between predicted age and true age, termed delta age, and used remaining time until death as our primary outcome to compare with. For some individual age groups (24, 34.5 and 36 months), delta age did indeed have a negative correlation with survival, with “younger” mice (those with a negative delta age) living longer at each individual age than “older” mice (those with a positive delta age) (Figure 1f, Table 1). For other groups this correlation is a trend, and more power may detect an association (Table 1). This suggests that the FI score is able to detect variation in predicted chronological age for mice of the same actual age, and this may represent biological age.
Table 1. Correlation coefficients (r2) and p values for correlation between survival and delta age determined by either FI score or FRIGHT age, at individual ages.
Figure 1. Frailty correlates with and is predictive of age in mice.
(a) Kaplan-Meier survival curve for male C57BL/6 mice (n=51) assessed longitudinally for frailty index (indicated by arrows). (b) Box and whisker plots for mean frailty index (FI) score for mice from 21 to 36 months of age. © FI score trajectories for each individual mouse from 21 months until death. (d) Univariate regression of FI score for chronological age on a training dataset, and (e) a testing dataset. (f) Residuals of the regression (delta age), plotted against survival for individual ages. Regression lines are only graphed for ages where there is an r2 value > 0.1.
Individual frailty items vary in their correlation with age
While a simple linear regression on overall frailty score was somewhat predictive of age, we hypothesized that by differentially weighting individual metrics, we could build a more predictive model, as has been done with various CpG sites to build methylation clocks (Horvath 2013b). To this end, we calculated the correlation between each individual frailty index item and chronological age (Table 2). Some parameters, such as tail stiffening, breathing rate/depth, gait disorders, hearing loss, kyphosis, and tremor, are strongly correlated (r2 > 0.35, p < 1e-30) with age (Figure 2), while others show very weak or no correlation with age (Table 2, Supplementary Figure 2). The fact that some parameters were very well correlated and others poorly correlated suggested that by weighting items we could build an improved model for biological age prediction.
Table 2. Correlation coefficients (r2) and p values for individual frailty items with chronological age.
Figure 2. Individual FI items vary in their correlation with age.
Mean scores across all mice (black line) for the top nine individual items of the frailty index that were correlated with chronological age. Colors indicate proportion of mice at each age with each score (0, blue; 0.5, orange, 1, red).
Multivariate regressions of individual frailty items to predict age (FRIGHT age)
We compared FI score as a single variable and four types of multivariate linear regression models to predict chronological age: simple least squares regression, elastic net regression, random forest regression and the Klemera-Doubal biological age estimation method (Klemera and Doubal 2006). We employed the bootstrap method on the training dataset to compare models. Only frailty items that had a significant, even if weak, correlation with age (p<0.05) were included in the analysis (see Table 2). The multivariate models, particularly elastic net, the Klemera-Doubal method and random forest were superior to FI as a single variable, with lower median error (p<0.0001, F=50.46, df=495), higher r2 values (p<0.0001, F=74.38, df=496), and smaller p-values (p<0.0001, F=33.43, df=495) when compared with one-way ANOVA. For further analysis, we selected the random forest regression model as it had the lowest median error (Figure 3a-c). Random forest models can also represent complex interactions among variables, which linear regressions cannot do, and may perform better in datasets where the number of features approaches or exceeds the number of observations (Breiman 2001). We term the outcome of this model FRIGHT age for Frailty Inferred Geriatric Health Timeline.

Also tagged with one or more of these keywords: machine learning, mice, aging

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users