AI-based investigation of molecular biomarkers of longevity
- PDF / 576,867 Bytes
- 14 Pages / 547.087 x 737.008 pts Page_size
- 13 Downloads / 190 Views
(0123456789().,-volV) ( 01234567 89().,-volV)
RESEARCH ARTICLE
AI-based investigation of molecular biomarkers of longevity Ihor Kendiukhov
Received: 25 April 2020 / Accepted: 30 June 2020 Ó Springer Nature B.V. 2020
Abstract In this paper, I build deep neural networks of various structures and hyperparameters in order to predict human chronological age based on openaccess biochemical indicators and their specifications from the NHANES database. In total, 1152 neural networks are trained and tested. The algorithms are trained and tested on incomplete data: missing values in data records are extrapolated by mean or median values for each parameter. I select the best neural networks in terms of validation accuracy (coefficient of determination and mean absolute error). It turns out that the most accurate results are delivered by multilayer networks (6 layers) with recurrent layers. Neural network types are selected by trial and error. The algorithms reached an accuracy of 78% in terms of coefficient of determination and 6.5 in terms of mean absolute error. I also list empirically determined features of neural networks that increase accuracy for the task of chronological age prediction. Obtained results can be considered as an approximation of human biological age. Parameters in training datasets are selected the most broadly: all potentially relevant parameters (926) from the NHANES database are I. Kendiukhov (&) School of Business and Economics, Humboldt University of Berlin, Unter den Linden 6, 10099 Berlin, Germany e-mail: [email protected] I. Kendiukhov Faculty of Biology, Zaporizhzhia National University, Zhukovskogo st., 10, Zaporizhzhia 69600, Ukraine
used. Although the networks are trained on the incomplete data, they demonstrated the ability to make reasonable predictions (with R2 [ 0.7) based on no more than 100 biochemical indicators. Hence, for practical reasons the full data on each of 926 indicators are not required, although the analysis of the impact of each indicator is useful for theoretical developments. Keywords Longevity biomarkers Deep neural networks Age prediction AI in biogerontology Machine learning
Introduction Machine learning algorithms are increasingly used in the analysis of biomarkers of human aging. Deep neural networks have proven to be able to find complex hidden patterns and regularities in the impact of biochemical indicators on human age. Various aging biomarkers panels are being proposed. By creating an algorithm for determining a person’s chronological age using molecular biomarkers, we also create an algorithm for estimating a person’s biological age, since on average, biological age for a population, by definition, is proportional to the chronological age. For a detailed explanation of the ability of machine learning algorithms trained on chronological age to predict biological age, see
123
Biogerontology
‘‘Discussion’’ section. Under quite general assumptions, such algorithms will predict biological age even more accurately than chronological one. Nevert
Data Loading...
