The Omics to Phenomics model generator. An AI/ML tool that learns patterns of biomarker combinations in omics datasets towards the development of predictive models with bespoke sensitivity and specificity. Applies ML using a proprietary genetic/evolution-inspired algorithm.
The Omics to Phenotype Prediction tool. A software tool that enables the generation of a particular phenotype classification based on unknown data (e.g. gene expression levels) and using a model generated by O2Pmgen. The tool generates a phenotype score and the respective phenotype classification (e.g. positive/negative).
Tool for the generation of the Reciever Operating Characteristic (ROC) curve of a model based on a given validation dataset. Computes and plots the ROC values of sensitivity and 1-specificity with higher precision based on a deterministic approach. Provides the calculation of the ROC Area Under the Curve (AUC) for accessing the model's predictive power.
The False Discovery Rate Analyser. A tool for the analysis of the distribution of false positives, false negatives, true positives and true negatives across the model scoring space. Enables the evaluation of models in terms of the likelihood of being correct for a given score. Generates a visualization that facilitates the evaluation and comparison of models.
The Biomarker Finder. A tool that enables a systematic search of putative biomarkers with statistical significance on large datasets. The tool enables testing numerous biomarker candidates using systematic hypothesis testing. Returns a list of candidates with associated p-values, expected alteration and estimated sensitivity and specificity as a single biomarker predictor.
The Data to Set generator. A tool that facilitates the construction of datasets from individual files. The tool enables a fast and fully automated dataset generation, compatible with ML tools such as O2Pmgen. Uses a sorting system and generates a dataset condensed file mapped with a metadata file with phenotype classification outcome.