Recapitulation of Ayurveda constitution types by machine learning of phenotypic traits

Abstract

In Ayurveda system of medicine individuals are classified into seven constitution types, “Prakriti”, for assessing disease susceptibility and drug responsiveness. Prakriti evaluation involves clinical examination including questions about physiological and behavioural traits. A need was felt to develop models for accurately predicting Prakriti classes that have been shown to exhibit molecular differences. The present study was carried out on data of phenotypic attributes in 147 healthy individuals of three extreme Prakriti types, from a genetically homogeneous population of Western India. Unsupervised and supervised machine learning approaches were used to infer inherent structure of the data, and for feature selection and building classification models for Prakriti respectively. These models were validated in a North Indian population. Unsupervised clustering led to emergence of three natural clusters corresponding to three extreme Prakriti classes. The supervised modelling approaches could classify individuals, with distinct Prakriti types, in the training and validation sets. This study is the first to demonstrate that Prakriti types are distinct verifiable clusters within a multidimensional space of multiple interrelated phenotypic traits. It also provides a computational framework for predicting Prakriti classes from phenotypic attributes. This approach may be useful in precision medicine for stratification of endophenotypes in healthy and diseased populations.

Genetic differences between extreme and composite constitution types from whole exome sequences reveal actionable variations

“…This observation is important for biological understanding of Vata and Kapha functions reflected in Prakriti features governed by their dry and humid nature respectively. There are specific therapeutic strategies described in Ayurveda for restoring Vata-Kapha balance through the management of fluid balance [28,29]. This genotype-phenotypic association though extremely preliminary, provides a novel window to explore water balance disorders.…”

“…This include 108 (18 x 3 x 2) individuals with 18 each Vata (V), Pitta (P) and Kapha (K) in each cohort. An extensive protocol was followed for recruitment of subjects, clinical phenotyping, classification into predominant groups as well as establishment of genetic homogeneity as previously published [28,29].…”

“…They display extreme phenotypic variations, with highly contrasting drug responses and disease susceptibilities [25,27]. Unsupervised machine learning and advanced statistical approaches on phenotypes of healthy individuals of extreme Prakriti types have been used to validate the existence of Prakriti specific phenotype-phenotype connectivity [29]. Studies have also provided evidence for differences at different hierarchies of genetic, epigenetic, biochemical and microbiome between Prakriti types [28][29][30][31][32][33].…”

“…Unsupervised machine learning and advanced statistical approaches on phenotypes of healthy individuals of extreme Prakriti types have been used to validate the existence of Prakriti specific phenotype-phenotype connectivity [29]. Studies have also provided evidence for differences at different hierarchies of genetic, epigenetic, biochemical and microbiome between Prakriti types [28][29][30][31][32][33]. Using this approach, we have previously identified predictive markers in EGLN1, a gene linked to high altitude adaptation [32,34].…”

Adhatoda Vasica rescues the hypoxia dependent severe asthma symptoms and mitochondrial dysfunction

“…Vata Pitta and Kapha. Tridosha are three physiological entities which govern the functioning and homeostasis of the system, which have been shown to have the molecular correlation (22, 23, [39][40][41][42]. The drugs prescribed in Ayurveda are also classified on the basis of their effect on Tridosha (43).…”