Biomarker discovery by sparse canonical correlation analysis of complex clinical phenotypes of tuberculosis and malaria.

Biomarker discovery aims to find small subsets of relevant variables in 'omics data that correlate with the clinical syndromes of interest. Despite the fact that clinical phenotypes are usually characterized by a complex set of clinical parameters, current computational approaches assume univariate targets, e.g. diagnostic classes, against which associations are sought for. We propose an approach based on asymmetrical sparse canonical correlation analysis (SCCA) that finds multivariate correlations between the 'omics measurements and the complex clinical phenotypes. We correlated plasma proteomics data to multivariate overlapping complex clinical phenotypes from tuberculosis and malaria datasets. We discovered relevant 'omic biomarkers that have a high correlation to profiles of clinical measurements and are remarkably sparse, containing 1.5-3% of all 'omic variables. We show that using clinical view projections we obtain remarkable improvements in diagnostic class prediction, up to 11% in tuberculosis and up to 5% in malaria. Our approach finds proteomic-biomarkers that correlate with complex combinations of clinical-biomarkers. Using the clinical-biomarkers improves the accuracy of diagnostic class prediction while not requiring the measurement plasma proteomic profiles of each subject. Our approach makes it feasible to use omics' data to build accurate diagnostic algorithms that can be deployed to community health centres lacking the expensive 'omics measurement capabilities.

Discriminating active from latent tuberculosis in patients presenting to community clinics.

BACKGROUND: Because of the high global prevalence of latent TB infection (LTBI), a key challenge in endemic settings is distinguishing patients with active TB from patients with overlapping clinical symptoms without active TB but with co-existing LTBI. Current methods are insufficiently accurate. Plasma proteomic fingerprinting can resolve this difficulty by providing a molecular snapshot defining disease state that can be used to develop point-of-care diagnostics. METHODS: Plasma and clinical data were obtained prospectively from patients attending community TB clinics in Peru and from household contacts. Plasma was subjected to high-throughput proteomic profiling by mass spectrometry. Statistical pattern recognition methods were used to define mass spectral patterns that distinguished patients with active TB from symptomatic controls with or without LTBI. RESULTS: 156 patients with active TB and 110 symptomatic controls (patients with respiratory symptoms without active TB) were investigated. Active TB patients were distinguishable from undifferentiated symptomatic controls with accuracy of 87% (sensitivity 84%, specificity 90%), from symptomatic controls with LTBI (accuracy of 87%, sensitivity 89%, specificity 82%) and from symptomatic controls without LTBI (accuracy 90%, sensitivity 90%, specificity 92%). CONCLUSIONS: We show that active TB can be distinguished accurately from LTBI in symptomatic clinic attenders using a plasma proteomic fingerprint. Translation of biomarkers derived from this study into a robust and affordable point-of-care format will have significant implications for recognition and control of active TB in high prevalence settings.