JADBio AutoML - State-of-the-art Automated Machine Learning Platform Software

You start by generating all your study data. That could either be data that has been processed and normalized by a Bioinformatician, or public data available in the known data repositories. You JUST upload that curated dataset to the JADBio app in a .csv or other delimited file format, select the desired predictive outcome, and watch the tool perform the analysis. Most analyses complete in a couple of hours.

• Disease Status (Diagnosis)
• Disease Subtype
• Response to Treatment
• Phenotypic Trait
• Time to Event (Death, Metastasis, Relapse)
• Other discrete or continuous quantities

• DNA
• Metabolites
• Clinical
• RNA
• Single Cell (any type)
• Signals (EEGs, etc.)
• Protein
• Sequence Data (SNP, methylation, etc.)
• Images

• Predictive models that could also be applied on new data
• A set of most-relevant to your question predictive biomarkers (biosignatures)
• Many visualizations to interpret results
• Decision support information to apply model

JADBio identifies accurate predictive models that use as few biomarkers as possible.

• Achieves an unprecedented level of automation in performing AI-based analysis
• Focuses on knowledge discovery applying novel feature selection algorithms
• Keeps getting better with usage employing meta-level learning, i.e., ML on the ML results
• Can also handle molecular data (low sample, high dimensional)
• Simultaneously handles different data types (multi-omics, genetic, medical images, medical signals, etc.)
• Correct and unbiased methodology
• Bio-specific functionalities, featuring Survival analysis, bio-specific data preprocessing
• Deep tech, Novel tech. Not just connecting existing black-box algorithms; developing new algorithms to solve open problems