Enlightening discriminative network functional modules behind Principal Component Analysis separation in differential-omic science studies

Abstract

Omic science is rapidly growing and one of the most employed techniques to explore differential patterns in omic datasets is principal component analysis (PCA). However, a method to enlighten the network of omic features that mostly contribute to the sample separation obtained by PCA is missing. An alternative is to build correlation networks between univariately-selected significant omic features, but this neglects the multivariate unsupervised feature compression responsible for the PCA sample segregation. Biologists and medical researchers often prefer effective methods that offer an immediate interpretation to complicated algorithms that in principle promise an improvement but in practice are difficult to be applied and interpreted. Here we present PC-corr: a simple algorithm that associates to any PCA segregation a discriminative network of features. Such network can be inspected in search of functional modules useful in the definition of combinatorial and multiscale biomarkers from multifaceted omic data in systems and precision biomedicine. We offer proofs of PC-corr efficacy on lipidomic, metagenomic, developmental genomic, population genetic, cancer promoteromic and cancer stem-cell mechanomic data. Finally, PC-corr is a general functional network inference approach that can be easily adopted for big data exploration in computer science and analysis of complex systems in physics.

Ceruloplasmin replacement therapy ameliorates neurological symptoms in a preclinical model of aceruloplasminemia

“…Multivariate analysis, which considers together the features collected for describing the brain pathophysiological state of each mouse, was done using unsupervised and parameter‐free (the algorithms do not require the tuning of any internal parameters, and hence, they prevent bias and overfitting) machine learning algorithms for linear and nonlinear dimension reduction. We applied principal component analysis (PCA), which is a state‐of‐the‐art approach for linear dimension reduction (Ringner, ; Ciucci et al , ) and minimum curvilinear embedding (MCE), which is a nonlinear and parameter‐free kernel PCA whose efficacy was extensively tested in previous studies (Cannistraci et al , , ; Ammirati et al , ; Alanis‐Lobato et al , ; Alessio & Cannistraci, ; Sales et al , ). We adopted both PCA and MCE to have confirmation that the results displayed were substantiate regardless of the type of linear or nonlinear transformation employed for their analysis.…”

Machine learning pattern recognition and differential network analysis of gastric microbiome in the presence of proton pump inhibitor treatment orHelicobacter pyloriinfection

“…Furthermore, we investigated the effect of PPI on the microbiota of gastric fluid and gastric mucosa in dyspeptic patients, and the changes induced by H. pylori infection on the gastric mucosal microbiota, by means of the PC-corr approach 62 . PC-corr represents a simple algorithm that associates to any PCA segregation a discriminative network of features’ interactions 62 .…”

Cell Mechanics Based Computational Classification of Red Blood Cells Via Machine Intelligence Applied to Morpho-Rheological Markers

“…We adopted PCA, which is a machine learning method for unsupervised linear and parameter-free dimension reduction. We performed unsupervised analysis instead of a supervised one, because it is less prone to overfitting as shown in previous studies [12], [17], [18]. 10,000 resampled datasets were generated from the original training set (P1-Partition 1), each of which was obtained by randomly selecting 200 mRBCs and 200 RETs.…”

“…In fact, the area is by definition a function of x-size and y-size, and PC-corr successfully infers this from the data independently from the single discriminative power of each feature. This result is possible because PC-corr is a multivariate approach and offers results different from univariate analysis approaches (which test single features), as extensively discussed in the article of Ciucci et al [12].…”