> For the complete documentation index, see [llms.txt](https://docs.neurolabai.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.neurolabai.io/readme.md). # Welcome to Neurolab AI Designed to accelerate agentic drug discovery pipelines in the Bio ecosystem and beyond, NeuroLab AI uses a proprietary collection of neural networks to predict the cognitive, molecular, and functional effects of any known or novel chemical compound on the human brain. Trained on a state-of-the-art biomedical knowledge graph consisting of 150,000+ nodes and over 8 million edges, NeuroLab AI produces data-rich, comprehensive human and machine-readable output encompassing 400+ neuroanatomical regions, and \*. \ \ Furthermore, Neurolab unites compound queries with machine-learned rankings, evidence inspection, natural-language data exploration, knowledge-graph context, and exploratory brain views into one application. Instead of producing single verdict, Neurolab gives researchers a structured environment to investigate a compound, inspect how a result was produced, follow the available support, compare possibilities, and see where the system is uncertain. In short, the Neurolab AI system provides an interactive research platform for compound-to-mechanism, evidence, and brain-system investigation. {% embed url="" %} {% hint style="warning" %} Disclaimer: Neurolab is a research and hypothesis-generation system. It is not medical advice, clinical decision support, a dosing guide, or a claim that every output is experimentally validated. {% endhint %} ## Choose a reading path To help you navigate the Neurolab documentation and its many sections, select from any of the reading paths below. \ \ Each reading path is organized in a specific sequence, each for different interests and experience levels. ### Path #1: I want to use the Neurolab application Start with [Run Your First Query](/start-using-neurolab/first-query.md), using **Donepezil** as the walkthrough compound. Then learn how to move between the report, [Section Inspector and Data Explorer](/start-using-neurolab/section-inspector-and-data-explorer.md), raw payload, report-grounded chat, [Knowledge Graph Explorer](/start-using-neurolab/knowledge-graph-explorer.md), and [Brain Explorer](/start-using-neurolab/brain-explorer.md). Continue with the captured examples: [investigate a known compound](/examples-and-use-cases/known-compound-investigation.md), [trace one report claim](/examples-and-use-cases/trace-a-report-claim.md), use [Ketamine to explore a multi-system report](/examples-and-use-cases/ketamine-multisystem-investigation.md), and use [Modafinil to investigate uncertainty](/examples-and-use-cases/investigate-uncertainty.md). ### Path #2: I want to organize and compare research Explore the developing workflow for [Accounts and Preferences](/organize-and-compare-research/accounts-and-preferences.md), [Research Workbenches](/organize-and-compare-research/research-workbenches.md), [Molecular Portfolios](/organize-and-compare-research/molecular-portfolios.md), and [Data Studio](/organize-and-compare-research/data-studio.md). These pages describe the direction of the product and clearly label capabilities that are still in development. ### Path #3: I want to evaluate a Neurolab result Read [How to Read a Neurolab Report](/start-using-neurolab/report-interpretation.md), [Evidence States](/understand-results/evidence-states.md), and [Scores, Reliability, and Abstention](/understand-results/scores-reliability-and-abstention.md). These pages explain why a high rank is not automatically a probability, why a citation does not validate every nearby prediction, and why withholding a result can be the correct behavior. ### Path #4: I want to understand Neurolab's engineering Begin with [System Architecture](/engineering-and-scientific-assurance/system-architecture.md). Continue through [Model Inference](/engineering-and-scientific-assurance/model-inference-overview.md), [Knowledge Graph](/engineering-and-scientific-assurance/knowledge-graph.md), [Model Training](/engineering-and-scientific-assurance/model-training-overview.md), [Evaluation Methods](/engineering-and-scientific-assurance/evaluation-methods.md), and [Release State and Validation](/engineering-and-scientific-assurance/release-state-and-validation.md). Developers can use the separate [Neurolab API documentation](/developer-resources/api-documentation.md); no public API gateway is currently deployed. ### Path #5: I want to explore the Neurolab application's Crypto Mode To familiarize with the application's current and planned cryptocurrency features, start with the the [Crypto Mode Overview](/usdndt-the-neurolab-token/overview.md). This overview explains how the application's wallet awareness features and the `$NDT` coordination layer (planned and in development) fit alongside the core product without being required to use Neurolab. After the overview, continue to [Wallets and Supported Networks](/usdndt-the-neurolab-token/wallets-and-networks.md) and [$NDT and Neurolab: Commons](/usdndt-the-neurolab-token/ndt-and-neurolab-commons.md). ## Current Workflows At present, the most mature end-to-end workflow is a **known-compound investigation**: enter the name of a compound represented in the governed knowledge graph, run the known-compound model ensemble, and inspect the structured report with its route, release, reliability, and evidence metadata. Structure-first and receptor-conditioned workflows are also available as conditional capabilities that are currently in beta testing, with a distinct inference path and a separate validation boundary. These workflows are treated separately as compared to routes involving known compounds. ## A useful mental model ```mermaid flowchart LR Query["Compound query"] --> Route["Route selection"] Route --> Known["Known-compound model"] Route --> Structure["Structure / receptor-conditioned beta route"] Known --> Report["Structured research report"] Structure --> Report Report --> Views["Report · section inspection · raw payload · inquiry · graph · brain"] Views --> Review["Human interpretation and source review"] ``` Next: [What Does Neurolab Do? ](/introduction.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.neurolabai.io/readme.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.