Machine learning prediction of therapy response

Abstract

A method comprising: receiving, for each of a plurality of subjects, each having a specified type of cardiovascular or cardiometabolic disease and receiving at least one specified therapy from a set of therapies for treating cardiovascular and cardiometabolic diseases, a first score representing a first genetic predisposition in said subject to respond to one or more of said set of therapies; at a training stage, training a machine learning model on a training set comprising: (i) all of said first scores, and labels associated with a response in each of said subjects to said at least one specified therapy; and at an inference stage, apply said trained machine learning model to a target said first score received with respect to a target subject, to predict a response in said target subject to at least one of said therapies in said set.

What is claimed is: 
     
         1 . A system comprising at least one hardware processor and a non-transitory computer-readable storage medium having stored thereon program code, the program code executable by the at least one hardware processor to:
 obtain, from each of a plurality of subjects, a biological sample;   analyze the sample to identify at least one first single nucleotide polymorphism (SNP) associated with a trait of a specified type of a disease;   calculate a first polygenic score based on the identified at least one first SNP;   assess a statistical significance of the at least one first SNP with respect to said trait of the disease;   select from the identified at least one first SNP one or more second SNPs that affect a response to a therapy for treating the disease, said selection based on: (a) location in genes that affect a target of said therapy, and (b) the assessed statistical significance;   calculate a second polygenic score, based on the one or more second SNPs; and   at an inference stage, apply a trained machine learning model to the first polygenic score and the second polygenic score pertaining to a target subject of the plurality of subjects, to predict a response of said target subject to at least one therapy from a set of therapies.   
     
     
         2 . The system of  claim 1 , wherein the at least one hardware processor is configured to, at a training stage, train a machine learning model on a training set, wherein said training set comprises (a) the first polygenic score and the second polygenic score, and (b) labels associated with a response in each of said subjects to at least one therapy. 
     
     
         3 . The system of  claim 1 , wherein said second score is calculated, based, at least in part, on identifying the one or more second SNPs in said target subject. 
     
     
         4 . The system of  claim 1 , wherein said machine learning model is trained on a training set that comprises: (a) said first polygenic score; (b) said second polygenic score; and (c) labels associated with a response in each of said plurality of subjects to said at least one therapy, wherein said labels comprise a positive set, corresponding to first selected subjects who have received a specified drug treatment and met a criterion of response to said therapy, and a negative set, corresponding to second selected subjects who have received the drug treatment but failed to meet the criterion. 
     
     
         5 . The system of  claim 4 , wherein said trait is blood pressure risk, and wherein said first score is a polygenic score calculated, based, at least in part, on identifying said first SNP in said target subject. 
     
     
         6 . The system of  claim 1 , wherein said first score and said second score are calculated in real time, based, on parallel calculation per each chromosome. 
     
     
         7 . The system of  claim 1 , wherein selection of the one or more second SNPs is further based on location of the one or more second SNPs in a gene promoter region or in a gene enhancer region. 
     
     
         8 . The system of  claim 1 , wherein said set of therapies is selected from the group consisting of: Ace inhibitors; Angiotensin receptor lockers; Thiazides; Thiazide like; Beta blockers; Alpha antagonists; Alpha blockers; Vasodilators; Aldosterone antagonist; Renal denervation and Barostimulation Biguanides; Sulfonylureas; Meglitinide derivatives; Alpha-glucosidase inhibitors; Thiazolidinediones (TZDs); Glucagonlike peptide-1 (GLP-1) agonists; Dipeptidyl peptidase IV (DPP-4) inhibitors; Selective sodium-glucose transporter-2 (SGLT-2) inhibitors; Insulins; Amylinomimetics; Bile acid sequestrants; Dopamine agonists and bariatric surgery for Type 2 Diabetes; Antilipemic agents; Nicotinic acids; Bile sequestrants; Aspirin, Anti coagulants; Anti platelet; Nitrates; anti-inflammatory agents; DNA methyltransferase inhibitors; antiarrhythmic agents; cardiac and neurological interventions; heart transplants; Digoxin; Ionotropic agents; ivabradine; sacubitril/valsartan; histone deacetylase inhibitors; lifestyle regimens; dietary regimens; and physical activity regimens. 
     
     
         9 . The system of  claim 1 , wherein said trait is selected from a list consisting of: hypertension, systolic blood pressure (SBP), diastolic blood pressure (DBP), high blood glucose levels, diabetes, hypercholesterolemia, high lipids levels, coronary heart disease, heart failure, obesity, arrhythmia, and any combination thereof. 
     
     
         10 . The system of  claim 1 , wherein said response in each of said plurality of subjects and said target subject to said one or more specified therapies comprises at least one of: lowering of SBP, lowering of DBP, lowering of blood glucose levels, lowering of blood lipids levels, improving of coronary heart disease symptoms, improving of heart failure symptoms, lowering weight, and improving of arrhythmia symptoms. 
     
     
         11 . The system of  claim 1 , wherein said training set further comprises, with respect to at least some of said plurality of subjects, labels associated with clinical data selected from the group consisting of: SBP, DBP, age, sex, body mass index (BMI), diet parameters, cholesterol parameters, co-morbidities, physical activity parameters, family history of cardiovascular disease and/or cardiometabolic disease, stress parameters, alcohol consumption parameters, and smoking and tobacco usage, lab results, imaging studies, ECG studies additional medication, response and adverse effect of medications, allergies, RNA profile, single cell RNA sequencing, metabolomics, microbiome, genomics, epigenomics, and MicroRNA. 
     
     
         12 . A method comprising:
 obtaining, from each of a plurality of subjects, a biological sample;   analyzing the sample to identify at least one first SNP associated with a trait of a specified type of a disease;   calculating a first polygenic score based on the identified at least one first SNP;   assessing a statistical significance of the at least one first SNP with respect to said trait of the disease;   selecting from the identified at least one first SNP one or more second SNPs that affect a response to a therapy for treating the disease, said selection based on: (a) location in genes that affect a target of said therapy, and (b) the assessed statistical significance;   calculating a second polygenic score, based on the one or more second SNPs; and   at an inference stage, applying a trained machine learning model to the first polygenic score and the second polygenic score pertaining to a target subject of the plurality of subjects, to predict a response of said target subject to at least one therapy from a set of therapies.   
     
     
         13 . The method of  claim 12 , further comprising: at a training stage, training a machine learning model on a training set, wherein said training set comprises (a) the first polygenic score and the second polygenic score, and (b) labels associated with a response in each of said subjects to at least one therapy. 
     
     
         14 . The method of  claim 12 , further comprising calculating said second score based, at least in part, on identifying the one or more second SNPs in said target subject. 
     
     
         15 . The method of  claim 12 , further comprising training said machine learning model on a training set that comprises: (a) said first polygenic score; (b) said second polygenic score; and (c) labels associated with a response in each of said plurality of subjects to said at least one therapy, wherein said labels comprise a positive set, corresponding to first selected subjects who have received a specified drug treatment and met a criterion of response to said therapy, and a negative set, corresponding to second selected subjects who have received the drug treatment but failed to meet the criterion. 
     
     
         16 . The method of  claim 15 , wherein said trait is a blood pressure risk, and wherein the method further comprises calculating said first score as a polygenic score based, at least in part, on identifying said first SNP in said target subject. 
     
     
         17 . The method of  claim 12 , wherein said first score and said second score are calculated in real time, based on parallel calculation per each chromosome. 
     
     
         18 . The method of  claim 12 , wherein selection of the one or more second SNPs is further based on location of the one or more second SNPs in a gene promoter region or in a gene enhancer region. 
     
     
         19 . The method of  claim 12 , wherein said set of therapies is selected from the group consisting of: Ace inhibitors; Angiotensin receptor lockers; Thiazides; Thiazide like; Beta blockers; Alpha antagonists; Alpha blockers; Vasodilators; Aldosterone antagonist; Renal denervation and Barostimulation Biguanides; Sulfonylureas; Meglitinide derivatives; Alpha-glucosidase inhibitors; Thiazolidinediones (TZDs); Glucagonlike peptide-1 (GLP-1) agonists; Dipeptidyl peptidase IV (DPP-4) inhibitors; Selective sodium-glucose transporter-2 (SGLT-2) inhibitors; Insulins; Amylinomimetics; Bile acid sequestrants; Dopamine agonists and bariatric surgery for Type 2 Diabetes; Antilipemic agents; Nicotinic acids; Bile sequestrants; Aspirin, Anti coagulants; Anti platelet; Nitrates; anti-inflammatory agents; DNA methyltransferase inhibitors; antiarrhythmic agents; cardiac and neurological interventions; heart transplants; Digoxin; Ionotropic agents; ivabradine; sacubitril/valsartan; histone deacetylase inhibitors; lifestyle regimens; dietary regimens; and physical activity regimens. 
     
     
         20 . The method of  claim 12 , wherein said trait is selected from a list consisting of: hypertension, SBP, DBP, high blood glucose levels, diabetes, hypercholesterolemia, high lipids levels, coronary heart disease, heart failure, obesity, arrhythmia, and any combination thereof.