Fully automated machine learning (AutoML) for predictive modeling is becoming a reality, giving rise to a whole new field. We present the basic ideas and principles of Just Add Data Bio (JADBio), an AutoML platform applicable to the low-sample, high-dimensional omics data that arise in translational medicine and bioinformatics applications. In addition to predictive and diagnostic models ready for clinical use, JADBio focuses on knowledge discovery by performing feature selection and identifying the corresponding biosignatures, i.e., minimal-size subsets of biomarkers that are jointly predictive of the outcome or phenotype of interest. It also returns a palette of useful information for interpretation, clinical use of the models, and decision making. JADBio is qualitatively and quantitatively compared against Hyper-Parameter Optimization Machine Learning libraries. Results show that in typical omics dataset analysis, JADBio manages to identify signatures comprising of just a handful of features while maintaining competitive predictive performance and accurate out-of-sample performance estimation.

RAF-kinase inhibitor protein (RKIP) is a well-established tumor suppressor that is frequently downregulated in a plethora of solid and hematological malignancies. RKIP exerts antimetastatic and pro-apoptotic properties in cancer cells, via modulation of signaling pathways and gene products involved in tumor survival and spread. Here we review the contribution of RKIP in the regulation of early metastatic steps such as epithelial–mesenchymal transition (EMT), migration, and invasion, as well as in tumor sensitivity to conventional therapeutics and immuno-mediated cytotoxicity. We further provide updated justification for targeting RKIP as a strategy to overcome tumor chemo/immuno-resistance and suppress metastasis, through the use of agents able to modulate RKIP expression in cancer cells.

Chronic stress is thought to be involved in the occurrence and progression of multiple diseases, via mechanisms that still remain largely unknown. Interestingly, key regulators of the stress response, such as members of the corticotropin-releasing-hormone (CRH) family of neuropeptides and receptors, are now known to be implicated in the regulation of chronic inflammation, one of the predisposing factors for oncogenesis and disease progression. However, an interrelationship between stress, inflammation, and malignancy, at least at the molecular level, still remains unclear. Here, we attempt to summarize the current knowledge that supports the inseparable link between chronic stress, inflammation, and colorectal cancer (CRC), by modulation of a cascade of molecular signaling pathways, which are under the regulation of CRH-family members expressed in the brain and periphery. The understanding of the molecular basis of the link among these processes may provide a step forward towards personalized medicine in terms of CRC diagnosis, prognosis and therapeutic targeting.

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.

Food safety organizations indicate the likelihood of constant human and animal exposure to mycotoxin mixtures as a possible negative public health impact. Risk assessment demonstrates that certain mycotoxins of Aspergillus and Penicillium spp. are toxic and hold a significant genotoxic efficacy at nanomolar concentrations. The aim of the current study was to investigate the potential cytogenetic effects of sterigmatocystin (STER), ochratoxin A (OTA) and citrinin (CTN) alone or in combination, at pM to μΜ concentrations, on the human hepatocellular cancer cell line Hep3B. MTT reduction, mitotic divisions, cell cycle delays and sister chromatid exchange rates (SCE) were determined as endpoints of metabolic activity, cytotoxicity, cytostaticity, and genotoxicity, respectively. All mycotoxin treatments induce SCE rates from 10⁻¹² M, while their cytotoxic and cytostatic potential varies. In PRI and MI assays, but not at MTT, STER alone or in combination with OTA + CTN appeared cytostatic and cytotoxic, even at 10⁻¹² M, while CTN alone and all other combinations displayed substantial cellular survival inhibition in doses ≥ 10⁻⁸ M. Co-administration of STER + OTA or STER + CTN in concentrations ≤ 10⁻¹ M, increased the MI and MTT activity, while it did not affect the PRI. Mycotoxin co-treatments revealed in general similar-to-additive or antagonistic genotoxic and cytotoxic effects. Our results for the first time describe that STER alone or in combination with OTA and/or CTN share a cytotoxic and cytogenetic potential even at picoMolar concentrations on human hepatoma cells in vitro.