FRIGHT age and AFRAID clocks

This webapp deploys machine learning-derived models to predict age and lifespan from the frailty assessment of mice aged 21 months or older. These clocks are based on random forest regressions, trained on a naturally aging population of male C57Bl6/NIA mice (Schultz, Kane, et al., 2020).

FRIGHT age (Frailty Inferred Geriatric Health Timeline) is a measure of a mouse's apparent chronological age.
The AFRAID clock (Analysis of Frailty and Death) is a prediction of a mouse’s life expectancy.

The clocks can be run on this webapp by uploading your frailty data in the box on the left, or on users' computers in Python with the models available on our repository.

Development of frailty-based clocks: (a) Box and whisker plots for frailty index (FI) score for mice from 21 to 36 months of age. (b) Random forest regression of the individual frailty index items for chronological age on testing dataset (FRIGHT age). (c) Random forest regression of the individual frailty index items for life expectancy on testing datasets, plotted against actual survival by age group (AFRAID clock).

Using the clocks:

1. Measure the mouse clinical frailty index (FI) as described in the Scoring Fraily tab.
2. Calculate weight variables (recent and total weight change, threshold weight change, and body weight standard deviation) as described in Schultz et al., 2020, or to your closest approximation.
3. Put your data in the same format as the example datasheet, with each column representing an FI variable and each row representing an individual mouse/assessment. The variable names must exactly match the names in the example datasheet. If group names are included, the sheet must be sorted by group. The models will not run with missing data (see the Note below for dealing with missing datapoints).
4. Upload your datasheet on left using the "Choose File" and "Analyze" buttons. The resulting FRIGHT age and AFRAID scores for each mouse will be added to your datasheet, and can be downloaded using the "Download: Analyzed Data" link on the left. A preview of your data can be viewed under the Analyzed Data tab.

Note: If an error message is generated, check the following: (1) that your .csv file is formatted exactly the same as the example data; (2) if group names are included, that the file is sorted by group; and (3) that there are no missing values. If missing datapoints exist, we recommend substituting the missing value with the average value for age-matched mice in your cohort, as the best approximation.

A detailed description of each of the items included in the mouse clinical frailty index can be found in the following papers:

Whitehead, J.C., Hildebrand, B.A., Sun, M., Rockwood, M.R., Rose, R.A., Rockwood, K., and Howlett, S.E. (2014). A clinical frailty index in aging mice: comparisons with frailty index data in humans. J. Gerontol. A Biol. Sci. Med. Sci. 69, 621–632. link

Feridooni, H.A., Sun, M.H., and Howlett, S.E. (2015). Reliability of a frailty index based on the clinical assessment of health deficits in male C57BL/6J mice. J. Gerontol. A Biol. Sci. Med. Sci. 6, 686-693. link

Kane, A.E., Ayaz, O., Ghimire, A., Feridooni, H.A., and Howlett, S.E. (2017). Implementation of the mouse frailty index. Can. J. Physiol. Pharmacol. 10, 1149-1155. link

No data has been uploaded/analyzed.

Use the box on the left to select your frailty index data with the "Choose File" button. Then hit the "Analyze" button. A preview of your analyzed data can be viewed in this tab. You can download your data with the "Download: Analyzed Data" link in the box on the left.