Multi-omic assessment

Abstract

Described herein are methods such as multi-omic methods for assessing a disease such as cancer. The multi-omic methods may integrate proteomic, transcriptomic, genomic, lipidomic, or metabolomic data. The method screening diseases or disease states. Also described herein are methods for screening for diseases or disease states from biological samples. The methods may include assessing whether a nodule, mass, or cyst is cancerous.

1 . A multi-omic method, comprising:
 obtaining multi-omic data from a biofluid sample of a subject suspected of having cancer, the multi-omic data comprising proteomic measurements and lipid or metabolite measurements; and   applying a machine learning classifier to the multi-omic data to evaluate the cancer, wherein the classifier uses a combination of features comprising protein features and lipid or metabolite features, and distinguishes between biofluid samples of subjects with and without cancer using the combination of features with a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8.   
     
     
         2 . The method of  claim 1 , wherein obtaining multi-omic data comprises generating the proteomic measurements and lipid or metabolite measurements by mass spectrometry. 
     
     
         3 . The method of  claim 1 , wherein the proteomic measurements comprise at least 10 protein measurements, and the lipid or metabolite measurements comprise at least 10 lipid or metabolite measurements. 
     
     
         4 . The method of  claim 1 , wherein the lipid or metabolite measurements comprise lipid measurements, and wherein the lipid or metabolite features comprise lipid features. 
     
     
         5 . The method of  claim 1 , wherein the lipid or metabolite measurements comprise metabolite measurements, and wherein the lipid or metabolite features comprise metabolite features. 
     
     
         6 . The method of  claim 1 , wherein the lipid or metabolite measurements comprise lipid and metabolite measurements, and wherein the lipid or metabolite features comprise lipid and metabolite features. 
     
     
         7 . The method of  claim 1 , wherein the combination of features further comprises a clinical feature comprising age, gender, race, or smoking status. 
     
     
         8 . The method of  claim 1 , wherein the multi-omic data further comprises nucleic acid measurements, and the combination of features further comprises nucleic acid features. 
     
     
         9 . The method of  claim 8 , wherein the nucleic acid measurements comprise mRNA, microRNA, or methylation measurements, and nucleic acid features comprise mRNA, microRNA, or methylation features. 
     
     
         10 . The method of  claim 8 , wherein the nucleic acid measurements comprise mRNA, microRNA, and methylation measurements, and nucleic acid features comprise mRNA, microRNA, and methylation features. 
     
     
         11 . The method of  claim 1 , wherein the lipid or metabolite measurements comprise lipid and metabolite measurements, wherein the lipid or metabolite features comprise lipid and metabolite features, wherein the multi-omic data further comprises nucleic acid measurements comprising mRNA, microRNA, and methylation measurements, and wherein the classifier further comprises nucleic acid features comprising mRNA, microRNA, and methylation features. 
     
     
         12 . The method of  claim 1 , wherein obtaining multi-omic data comprises obtaining proteomic measurements from proteins of the biofluid sample adsorbed to nanoparticles. 
     
     
         13 . The method of  claim 1 , wherein obtaining multi-omic data comprises contacting the biofluid sample with internal standard proteins. 
     
     
         14 . The method of  claim 1 , wherein applying the classifier to the multi-omic data to evaluate the cancer comprises:
 applying a first classifier to the proteomic measurements to generate a first label corresponding to a presence, absence, or likelihood of the cancer,   applying a second classifier to the lipid or metabolite measurements to generate a second label corresponding to a presence, absence, or likelihood of the cancer, and   evaluating the cancer based on (a), (b) or (c):
 (a) a non-weighted average of the first and second labels, 
 (b) a weighted average of the first and second labels, or 
 (c) a majority voting score based on the first and second labels. 
   
     
     
         15 . The method of  claim 14 , further comprising evaluating the cancer based on the weighted average of the first and second labels, wherein the weighted average is generated by assigning weights to results of the first and second classifiers based on area under a ROC curve, area under a precision-recall curve, accuracy, precision, recall, sensitivity, F1-score, specificity, or a combination thereof. 
     
     
         16 . The method of  claim 1 , wherein applying the classifier to the multi-omic data to evaluate the cancer comprises:
 obtaining a subset of features from among the lipid or metabolite measurements;   obtaining at least a subset of features from among the proteomic measurements;   pooling the subset of features from among the lipid or metabolite measurements and the at least the subset of features from among the proteomic measurements to obtain pooled features; and   evaluating the cancer based on the pooled features.   
     
     
         17 . The method of  claim 16 , wherein obtaining the subset of features from among the lipid or metabolite measurements or obtaining at least the subset of features from among the proteomic measurements comprises obtaining top features based on univariate data. 
     
     
         18 . The method of  claim 1 , wherein applying the classifier to the multi-omic data to evaluate the cancer comprises:
 obtaining a subset of features from among the proteomic measurements;   obtaining at least a subset of features from among the lipid or metabolite measurements;   pooling the subset of features from among the proteomic measurements and the at least the subset of features from among the lipid or metabolite measurements to obtain pooled features; and   evaluating the cancer based on the pooled features.   
     
     
         19 . The method of  claim 18 , wherein obtaining the subset of features from among the proteomic measurements or obtaining at least the subset of features from among the lipid or metabolite measurements comprises obtaining top features based on univariate data. 
     
     
         20 . The method of  claim 1 , wherein the classifier is trained using deep learning, a random forest classification analysis, a support vector machine analysis, a naive Bayes analysis, or a hidden Markov analysis. 
     
     
         21 . The method of  claim 1 , wherein the classifier identifies the multi-omic data as indicative of cancer, and wherein the method further comprises administering a chemotherapy, pharmaceutical, radiation or surgical cancer treatment to the subject. 
     
     
         22 . The method of  claim 1 , wherein the cancer is selected from the group consisting of: lung cancer, pancreatic cancer, colon cancer, liver cancer, breast cancer, and ovarian cancer. 
     
     
         23 . The method of  claim 22 , wherein the cancer comprises early-stage cancer. 
     
     
         24 . The method of  claim 1 , wherein the biofluid sample comprises a blood, serum, or plasma sample. 
     
     
         25 . A multi-omic method, comprising:
 obtaining multi-omic data from a biofluid sample of a subject suspected of having cancer, the multi-omic data comprising lipid and metabolite measurements; and   applying a classifier to the multi-omic data to evaluate the cancer, wherein the classifier uses a combination of features comprising lipid and metabolite features, and distinguishes between biofluid samples of subjects with and without cancer using the combination of features with a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8.   
     
     
         26 . The method of  claim 25 , wherein obtaining multi-omic data comprises generating the lipid and metabolite measurements by mass spectrometry. 
     
     
         27 . The method of  claim 25 , wherein the lipid measurements comprise at least 10 lipid measurements, and the metabolite measurements comprise at least 10 metabolite measurements. 
     
     
         28 . A multi-omic method, comprising:
 obtaining multi-omic data from a biofluid sample of a subject suspected of having cancer, the multi-omic data comprising nucleic acid measurements and lipid or metabolite measurements; and   applying a classifier to the multi-omic data to evaluate the cancer, wherein the classifier uses a combination of features comprising nucleic acid features and lipid or metabolite features, and distinguishes between biofluid samples of subjects with and without cancer using the combination of features with a performance characterized by a receiver operating characteristic (ROC) curve having an average or median area under the curve (AUC) of at least 0.8.   
     
     
         29 . The method of  claim 28 , wherein obtaining multi-omic data comprises generating the nucleic acid measurements and lipid or metabolite measurements by mass spectrometry. 
     
     
         30 . The method of  claim 28 , wherein the nucleic acid measurements comprise at least 10 measurements, and the lipid or metabolite measurements comprise at least 10 measurements.