Precision medicine aims to move from traditional reactive medicine to a system where risk groups can be identified before the disease occurs. However, phenotypic heterogeneity amongst the diseased and healthy poses a major challenge for identification markers for risk stratification and early actionable interventions. In Ayurveda, individuals are phenotypically stratified into seven constitution types based on multisystem phenotypes termed “Prakriti”. It enables the prediction of health and disease trajectories and the selection of health interventions. We hypothesize that exome sequencing in healthy individuals of phenotypically homogeneous Prakriti types might enable the identification of functional variations associated with the constitution types. Exomes of 144 healthy Prakriti stratified individuals and controls from two genetically homogeneous cohorts (north and western India) revealed differential risk for diseases/traits like metabolic disorders, liver diseases, and body and hematological measurements amongst healthy individuals. These SNPs differ significantly from the Indo-European background control as well. Amongst these we highlight novel SNPs rs304447 (IFIT5) and rs941590 (SERPINA10) that could explain differential trajectories for immune response, bleeding or thrombosis. Our method demonstrates the requirement of a relatively smaller sample size for a well powered study. This study highlights the potential of integrating a unique phenotyping approach for the identification of predictive markers and the at-risk population amongst the healthy.
Multiple myeloma (MM) is a hematological disease marked by abnormal growth of B cells in bone marrow. Inherent chromosomal instability and DNA damage are major hallmarks of MM, which implicates an aberrant DNA repair mechanism. Studies have implicated a role for CDK12 in the control of expression of DNA damage response genes. In this study, we examined the effect of a small molecule inhibitor of CDK12–THZ531 on MM cells. Treatment of MM cells with THZ531 led to heightened cell death accompanied by an extensive effect on gene expression changes. In particular, we observed downregulation of genes involved in DNA repair pathways. With this insight, we extended our study to identify synthetic lethal mechanisms that could be exploited for the treatment of MM cells. Combination of THZ531 with either DNA-PK inhibitor (KU-0060648) or PARP inhibitor (Olaparib) led to synergistic cell death. In addition, combination treatment of THZ531 with Olaparib significantly reduced tumor burden in animal models. Our findings suggest that using a CDK12 inhibitor in combination with other DNA repair inhibitors may establish an effective therapeutic regimen to benefit myeloma patients.
Personalized medicine relies on successful identification of genome-wide variations that governs inter-individual differences in phenotypes and system level outcomes. In Ayurveda, assessment of composite constitution types “ Prakriti” forms the basis for risk stratification, predicting health and disease trajectories and personalized recommendations. Here, we report a novel method for identifying pleiotropic genes and variants that associate with healthy individuals of three extreme and contrasting “ Prakriti” constitutions through exome sequencing and state-of-the-art computational methods. Exome Seq of three extreme Prakriti types from 108 healthy individuals 54 each from genetically homogeneous populations of North India (NI, Discovery cohort) and Western India (VADU, Replication cohort) were evaluated. Fisher’s Exact Test was applied between Prakriti types in both cohorts and further permutation based p-value was used for selection of exonic variants. To investigate the effect of sample size per genetic association test, we performed power analysis. Functional impact of differentiating genes and variations were inferred using diverse resources -Toppfun, GTEx, GWAS, PheWAS, UK Biobank and mouse knockdown/knockout phenotype (MGI). We also applied supervised machine learning approach to evaluate the association of exonic variants with multisystem phenotypes of Prakriti . Our targeted investigation into exome sequencing from NI (discovery) and VADU (validation) cohorts datasets provide ~7,000 differentiating SNPs. Closer inspection further identified a subset of SNPs (2407 (NI) and 2393 (VADU)), that mapped to an overlapping set of 1181 genes. This set can robustly stratify the Prakriti groups into three distinct clusters with distinct gene ontological (GO) enrichments. Functional analysis further strengthens the potential pleiotropic effects of these differentiating genes/variants and multisystem phenotypic consequences. Replicated SNPs map to some very prominent genes like FIG4, EDNRA, ANKLE1, BCKDHA, ATP5SL, EXOCS5 , IFIT5, ZNF502, PNPLA3 and IL6R . Lastly, multivariate analysis using random forest uncovered rs7244213 within urea transporter SLC14A2 , that associate with an ensemble of features linked to distinct constitutions. Our results reinforce the concept of integration of Prakriti based deep phenotypes for risk stratification of healthy individuals and provides markers for early actionable interventions.