JADBio Platform Documentation
Welcome to the Documentation for Just Add Data Bio (JADBio). This site contains information on how to access the JADBio platform, use its interface and explore its API functionalities.
Overview
JADBio is a platform designed specifically to extract value and insights from life-science datasets (clinical, molecular, images, text and combinations thereof). The dataset to analyze is input to JADBio as a 2D matrix with the rows corresponding to samples and the columns corresponding to measured features (a.k.a. biomarkers, variables, measured quantities, predictors) and a defined outcome of interest. The AutoML system does not care if these features are amplicons, genes, proteins, etc or whether the samples are patients or single cells. JADBio can typically handle datasets with feature sizes, up to 1M and sample sizes, from 15 (per class) to 200K. In addition, JADBio accepts dichotomous (binary classification), nominal (multi-class classification), continues (regression), and time-to-event (e.g., survival analysis) type of outcomes.
JADBio's uniquely Automated Machine Learning (AutoML) is guided by Artificial Intelligence (AI) to provide accurate and efficient predictive models for Classification, Regression and Survival (Time to Event) analysis. Typical prediction tasks are (but not limited to)
- Disease Status (Diagnosis)
- Disease Subtype
- Response to Treatment
- Phenotypic Trait
- Time to Event (Death, Metastasis, Relapse)
You can visit JADBio and checkout our tool!
JADBio Platform Documentation contents
This documentation site is organized as follows:
- Getting Started
- Accessing JADBio
- JADBio's main screen
- JADBio Use Cases
- Starting a project
- Adding Data
- Prepare your data for JADBio
- Uploading data
- Step 1 & 2 - How to upload a file
- Step 3 - Setting up Rows & Columns layout
- Step 4 - Writing a name and a description for your dataset
- Attaching data
- Step 1 - Select how to add data
- Step 2 - Select a dataset to attach to your project
- Step 3 - Assign name and description
- Reviewing project data
- Data Wrangling
- Setting up an Analysis
- Starting an analysis
- Select dataset
- Previewing my uploaded dataset
- Analysis setup
- Selecting the outcome depending on the task
- Notifications and warnings before continuing (sample size, class distribution, etc)
- Default analysis options
- Advanced options
- Running analysis progress view
- Viewing Analysis Results
- Applying Generated Models
- Validating your model
- Choosing the best model and signature of interest
- Validate against labeled samples
- Predict the outcome for unlabeled samples
- Manually enter values to predict
- Try model outside JADBio (download model)
- Exporting models
- Account Maintenance