Interact with tabular or single-cell data using natural language, without requiring any coding skills. Ask questions, explore insights, visualize data, and generate statistics using a conversational interface, built on a state-of-the-art language model that can generate human-like text from provided inputs.

Example use cases: (i) data extraction, to obtain specific information such as summary statistics, correlations, outliers or trends; (ii) generating charts or graphs to visualize your data; (iii) performing simple data modelling or data analysis; (iv) data quality control, (v) training supervised or unsupervised machine learning models.

Technology: Large Language Models (LLMs)

Limitations: (i) ExploreBio may not always produce correct or efficient code. Outputs should always be reviewed before being published elsewhere; (ii) ExploreBio may not always understand the context, intent, or goal of your query. It may generate code that is irrelevant, incomplete, or inappropriate for your task. You should always provide clear and specific instructions, and use follow-up questions or feedback to refine the code if needed; (iii) code will time out if a task does not complete within 120 seconds--this means that more complicated analysis or machine learning may not complete in the allowed timeframe; (iv) for tabular data only the first 100 columns will be analysed, while for single cell data the analysis is carried out holistically; (v) a maximum of 8gb of RAM can be used by a single session.