Alzheimer’s disease (AD) is the most common form of neurodegenerative dementia and its timely diagnosis remains a major challenge in biomarker discovery. In the present study, we analyzed publicly available high-throughput low-sample -omics datasets from studies in AD blood, by the AutoML technology Just Add Data Bio (JADBIO), to construct accurate predictive models for use as diagnostic biosignatures. Considering data from AD patients and age–sex matched cognitively healthy individuals, we produced three best performing diagnostic biosignatures specific for the presence of AD: A. A 506-feature transcriptomic dataset from 48 AD and 22 controls led to a miRNA-based biosignature via Support Vector Machines with three miRNA predictors (AUC 0.975 (0.906, 1.000)), B. A 38,327-feature transcriptomic dataset from 134 AD and 100 controls led to six mRNA-based statistically equivalent signatures via Classification Random Forests with 25 mRNA predictors (AUC 0.846 (0.778, 0.905)) and C. A 9483-feature proteomic dataset from 25 AD and 37 controls led to a protein-based biosignature via Ridge Logistic Regression with seven protein predictors (AUC 0.921 (0.849, 0.972)). These performance metrics were also validated through the JADBIO pipeline confirming stability. In conclusion, using the automated machine learning tool JADBIO, we produced accurate predictive biosignatures extrapolating available low sample -omics data. These results offer options for minimally invasive blood-based diagnostic tests for AD, awaiting clinical validation based on respective laboratory assays. They also highlight the value of AutoML in biomarker discovery.

Molecular gene-expression datasets consist of samples with tens of thousands of measured quantities (i.e., high dimensional data). However, lower-dimensional representations that retain the useful biological information do exist. We present a novel algorithm for such dimensionality reduction called Pathway Activity Score Learning (PASL). The major novelty of PASL is that the constructed features directly correspond to known molecular pathways (genesets in general) and can be interpreted aspathway activity scores. Hence, unlike PCA and similar methods, PASL’s latent space has a fairly straightforward biological interpretation. PASL is shown to outperform in predictive performance the state-of-the-art method (PLIER) on two collections of breast cancer and leukemia gene expression datasets. PASL is also trained on a large corpus of 50000 gene expression samples to construct a universal dictionary of features across different tissues and pathologies. The dictionary validated on 35643 held-out samples for reconstruction error. It is then applied on 165 held-out datasets spanning a diverse range of diseases. The AutoML tool JADBio is employed to show that the predictive information in the PASL-created feature space is retained after the transformation. The code is available athttps://github.com/mensxmachina/PASL.

Background/Objectives: Alterations in DNA damage repair mechanisms can impair the therapeutic effectiveness of cisplatin. MicroRNAs (miRNAs), key regulators of DNA damage repair processes, have been proposed as promising biomarkers for predicting the response to platinum-based chemotherapy (CT) in non-small cell lung cancer (NSCLC). In this study, by using a bioinformatics approach, we identified six miRNAs, which were differentially expressed (DE) between NSCLC patients characterized as responders and non-responders to platinum-based CT. We further validated the differential expression of the selected miRNAs on tumor and matched normal tissues from patients with resected NSCLC. Methods: Two miRNA microarray expression datasets were retrieved from the Gene Expression Omnibus (GEO) repository, comprising a total of 69 NSCLC patients (N = 69) treated with CT and annotated data from their response to treatment. Differential expression analysis was performed using the Linear Models for Microarray Analysis (Limma) package in R to identify DE miRNAs between responders (N = 33) and non-responders (N = 36). Quantitative real-time PCR (qRT-PCR) was used to assess miRNA expression levels in clinical tissue samples (N = 20). Results: Analysis with the Limma package revealed 112 DE miRNAs between responders and non-responders. A random-effects meta-analysis further identified 24 miRNAs that were consistently up- or downregulated in at least two studies. Survival analysis using the Kaplan–Meier plotter (KM plotter) indicated that 22 of these miRNAs showed significant associations with prognosis in NSCLC. Functional and pathway enrichment analysis revealed that several of the identified miRNAs were linked to key pathways implicated in DNA damage repair, including the p53, Hippo, PI3K and TGF-β signaling pathways. We finally distinguished a six-miRNA signature consisting of miR-26a, miR-29c, miR-34a, miR-30e-5p, miR-30e-3p and miR-497, which were downregulated in non-responders and are involved in at least three DNA damage repair pathways. Comparative expression analysis on tumor and matched normal tissues from surgically treated NSCLC patients confirmed their differential expression in clinical samples. Conclusions: In summary, we identified a signature of six miRNAs that are suppressed in NSCLC and may serve as a predictor of cisplatin response in NSCLC.