> ## Documentation Index
> Fetch the complete documentation index at: https://docs.purna.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Key concepts

> Understand the core building blocks of Purna.

Purna is a chat-first molecular intelligence platform for biological research. It brings reasoning, scientific databases, files, code execution, background jobs, and structured outputs into one workspace.

## Chat

Chat is the primary way you interact with Purna. Submit natural-language requests and receive responses with tool traces, citations, tables, code, visualizations, or downloadable artifacts. Conversations are saved automatically and appear in the sidebar.

## Research modes

Research modes shape the assistant's prompt, tool choices, and suggested workflow.

| Mode                    | Best for                                                                                           |
| ----------------------- | -------------------------------------------------------------------------------------------------- |
| **Biomedical Research** | Literature, pathways, disease mechanisms, genes, proteins, drugs, and general biological reasoning |
| **Omics Analysis**      | RNA-seq, single-cell, proteomics, metabolomics, multi-omics, and uploaded dataset analysis         |

Research mode is selected before a chat begins and is locked after the first message. Start a new chat to use a different mode.

## Preferences

The **Preferences** menu in the chat input controls focused workflows and tool availability.

* **Workflow modes** activate deeper experiences such as Deep Research, Consilium, and Paper Writing.
* **Capabilities** control tool families such as web search, code execution, image generation, and background jobs.
* **Skills and knowledge** control source-aware guidance and knowledge resources.

You can remember preferences so new chats start with the same tool configuration.

## Projects

Projects group related chats, files, jobs, and generated outputs. Use them for a study, literature review, protein design task, omics analysis, or team research direction. Each account starts with **Quick Research** as the default project.

## Files

Files are uploaded datasets, documents, images, and generated outputs. You can attach files to chat, upload them from the Files page, reference them with `@`, and use them in code execution or background jobs.

## Artifacts

Artifacts are structured outputs Purna creates during conversations:

* **Documents** for summaries, briefs, protocols, drafts, and reports
* **Code files** for reusable Python or R scripts
* **Spreadsheets** for tabular results and exports
* **Images and figures** for visual analysis outputs
* **Structure and protein viewers** for molecular visualization workflows

Artifacts appear in a side panel and can be revised through follow-up prompts.

## Code execution

Code execution runs Python or R on a private compute server attached to your chat. Use it for custom statistics, data transformation, visualizations, and exploratory analysis that needs real code.

## Jobs

Jobs are autonomous background workflows for longer tasks. They run in isolated sandboxes, can install packages, process files, generate outputs, and sync results back to your Files library.

## Database search

Purna can query scientific sources across literature, clinical evidence, variant databases, pathways, proteins, drugs, and web search. You can control available sources and capabilities through **Preferences**.

## Citations

Responses include citation badges and source panels when Purna uses external evidence. Click citations to inspect the source record and verify claims.

## Teams and visibility

Private chats remain visible only to you. Team-visible projects and shared chats let collaborators review selected work. Use team settings to invite members and manage roles.