Formaticon Documentation
Your complete guide for using the Cellformatica AI-powered tools for bioscience research - from submitting general unstructured tasks to AI scientist (write a paper, perform data analysis or literature research) to running preconfigured instruments (interpret a candidate target list or generate a rich analysis of a gene set ) .
What is Formaticon?
Formaticon is an AI-driven bioscience research platform developed by Cellformatica. It combines large language models with systematic literature retrieval, pathway analysis, and structured report generation to help researchers answer scientific questions at scale. Whether you need to screen hundreds of genes against a hypothesis, generate a multi-section analytical report, or commission a fully autonomous deep-research pipeline, Formaticon delivers publication-ready outputs, complete with code, figures, hypotheses, PRISMA-compliant search protocols, and formatted references.
Three ways to do research
Bioscientist Agent
Describe any open-ended research question or task. The autonomous pipeline searches, reasons, hypothesizes, and delivers a full PDF/DOCX white paper.
Batch Query
Screen a list of entities โ genes, drugs, diseases, cell types โ against a research question. Get an interactive ranked table and scatter plot.
Gene module analysis
Submit a focused gene list and research question. Receive a multi-section narrative report with regulatory network analysis, figures, and references.
Quick start
-
Create an account Register at formaticon.cellformatica.com/register with your name, email, phone, and a password. Then log in.
-
Add credits Each run consumes credits based on AI compute used. Open your Profile page and follow the billing link to purchase credits via Stripe.
-
Choose a workflow Use the table below or the navigation to choose: Batch Query for entity screening, Detailed Report for a curated gene set, or Bioscientist Agent for open-ended deep research.
-
Fill in the form and submit Enter your research question and any required inputs (entity list, gene list, or a detailed task description). Review the cost estimate, then confirm.
-
Wait for results Batch Queries typically complete in a few minutes. Detailed Reports take 30โ90 minutes. Bioscientist Agent tasks run for 1โ2.5 hours. You can monitor status in real time.
-
Download and share View your results in the browser, download PDF and DOCX files, export CSVs, and share results with colleagues via a secure access-token link.
Choosing the right workflow
| Criterion | Batch Query | Gene module analysis | Bioscientist Agent |
|---|---|---|---|
| Input | Question + entity list (genes, drugs, diseases, etc.) | Question + gene list (<30 genes) | Open-ended question or research task (+ optional files) |
| Output | Ranked table + scatter plot + CSV | PDF/DOCX multi-section report | PDF/DOCX white paper + figures + hypotheses + PRISMA |
| Runtime | A few minutes | 30โ90 minutes | 1โ2.5 hours |
| Best for | Screening many candidates against one question | Narrative causal analysis of a curated gene set | Systematic reviews, white papers, exploratory deep research |
| Downloads | CSV | PDF, DOCX | PDF, DOCX, figures, session log, RIS references |
In this documentation
- Getting StartedAccount creation, login, credits, and choosing a workflow
- Batch QueryForm fields, focusing terms, results table, scatter plot
- Detailed ReportGene list input, report structure, PDF/DOCX download
- Bioscientist AgentConversational agent, task submission, phase tracker, downloads
- Reading Your ResultsTable columns, scatter axes, white paper sections, PRISMA, hypotheses
- History, Sharing & BillingHistory pages, share links, credits, billing, NOT_ENOUGH_MONEY
- FAQCommon questions across all workflows