Bioinformatics AI
From FASTQ to Figures

Bench Scientists / Molecular Biologists.

These are researchers who routinely generate sequencing data (RNA-seq, ATAC-seq, scRNA-seq, etc.) but depend on a bioinformatics core or collaborator to analyze it. Pipette helps them gain time and independence through instant access to analysis using natural language:
"Run differential expression on my treated vs control samples."

Small Labs, Startups, and Core Facilities.

Small groups or startup labs often can’t afford a dedicated bioinformatician. Pipette serves as a “virtual core,” capable of automating routine workflows (QC, alignment, DE, enrichment) and documenting provenance. Pipette enables scalability without headcount.

Bioinformaticians and Computational Biologists

Pipette handles the environment setup, pipeline execution, and reproducibility tracking, letting experts focus on interpretation and custom modeling. For them, Pipette is a force multiplier: it automates what’s tedious, preserves what’s critical (metadata, versions), and allows scripting extensions for advanced use cases.

Principal Investigators and Project Leads

PIs get a unified, auditable record of all analyses, so when a postdoc leaves, the data and pipelines remain intact. The “Provenance Reports” also make grant renewals and publications more defensible. In other words: less chaos, more continuity.

Industry & Translational Teams

Biotech and pharma teams that need standardized, version-tracked workflows for regulatory or QA reasons benefit from Pipette’s provenance system and automated documentation. Every run is traceable, which simplifies QA audits and cross-team validation.

Pipette is designed to support a wide range of bioinformatics workflows across diverse areas of life science research. It integrates a broad suite of widely adopted open-source tools commonly used in bioinformatics and computational biology. Currently, Pipette operates with two active analysis engines, enabling robust support for various domains, including human genomics, plant and crop research, microbiome analysis, and metagenomics. Over time, we will add tools for metabolomics, drug-designing, proteomics. Beyond running pipelines, Pipette also generates hypotheses grounded in scientific literature and suggests follow-up computational experiments based on your results. This breadth makes Pipette a flexible, general-purpose platform for both exploratory and production-scale bioinformatics.

Every analysis records its full data lineage: inputs, parameters, software versions, outputs, and code. The rich provenance model can be utilized to reproduce results on any machine later.

No. Pipette runs entirely in your browser. Just log in, upload your data, and start your analysis through natural conversation.

Pipette currently supports a variety of popular bioinformatics tools for NGS analysis and computational biology. Please check the platform page to check the list of currently available tools.

Just use the feedback form to request the tool. If the tool passes our internal checks, we will make it available ASAP.

Follow these steps:

• Set your compute environment. Each compute engine is equipped with tools specific to that domain.
• Use the upload tool in the bucket page to upload a dataset. It accepts most bioinformatics file formats (fastq, h5, bam, csv etc). Wait for the upload to finish.
• In the home page, briefly describe your dataset and goals in the prompt box. It's best to let Pipette know your organism, data type, and what kind of files you want as output. Click send and wait for logs to appear in the progress window. At this point, its safe to close the browser or sign out. You can log back in and check progress in the tasks page.
• Once the task is finished, the reporting agent will read all output files and generate a detailed report on the reports window. Next, copy the 'task ID' to download output files generated by Pipette in the bucket page.
• Click in the bucket icon on the side bar and paste the task ID in the search bar. The session directory will appear in the window below. Click on the directory to inspect and download output files.
  
  You can change the name of output files and directory names (except the directory named 'outputs'). If you want to use any output files as input in the another session, drag and drop the file to the base directory (outside outputs). Placing any file here automatically becomes available as input to any session inside Pipette.bio. You can also move the file using the move tool. To move to base dir, type "/filename.txt" in the dialog box that opens on clicking the move tool.
• Use the recordings page to access previous tasks. Currently, Pipette cannot load the chat history into a new session, but you can use the output files from any previous session (as described below) and continue analyzing inside a new session.