Run bioinformatics pipelines without leaving the conversation

The handoff was the hard part

RNA-seq, ChIP-seq, variant calling. Preclinical teams generate this data constantly, and running the analysis was rarely the bottleneck. The bottleneck was everything around it: upload files to one system, queue jobs in another, hunt down QC reports in a third, then paste the results back into the thread where the actual decision is happening.

That fragmentation costs time. Worse, it breaks the thread between the question you asked and the pipeline that answered it.

Starting now, it doesn't have to.

What's new

EMET can run Galaxy bioinformatics pipelines directly from the conversation. Galaxy is backend compute you never have to log into. There's no separate server URL, no API key to manage, no job history to navigate in a browser.

When you're ready to analyze private data, attach a Bioinformatics File in the composer (FASTQ, FASTA, or GZ). EMET provisions a per user Galaxy account on first use, uploads the file into your workspace, and runs the tools or workflows you ask for against that data. You describe what you want in plain language. EMET handles provisioning, job submission, polling, and delivery. Results come back as artifacts in the turn: QC reports you can review inline, tables you can download, outputs ready for the next step.

How it works, in practice

Here we showcase two workflows, but there are countless possibilities.

Before alignment or quantification, you need to know whether your FASTQ files are usable: per base quality, adapter content, duplication, GC bias. Upload paired end reads, ask EMET to run FastQC, and the report comes back in chat. No standalone QC tool, no download and reattach loop.

Once reads are aligned and counted, the question shifts to which genes change between conditions. Galaxy's gene expression tools support full RNA-seq pipelines: alignment with STAR or Salmon, quantification with featureCounts, differential expression with DESeq2, edgeR, or limma-voom. Upload FASTQ files, or a count matrix if you already have one, describe the comparison (treated versus control, knockout versus wild type), and EMET runs the pipeline and returns normalized counts and differential expression tables.

These are the proven, high frequency paths. EMET's self hosted Galaxy instance also exposes a broader tool catalogue, variant calling, ChIP-seq, single-cell preprocessing, metagenomics among them, and the platform is built to extend into multi-step pipelines as programs need them.

Security and data governance

Each scientist gets their own Galaxy account, so jobs and histories don't mix between users. Credentials are provisioned and encrypted automatically. You never see or manage them. Analyses run on EMET's private, self hosted Galaxy, not on public mirrors, so your data stays in the environment your organization controls.

Why it matters

For scientists, this closes the gap between having sequencing data and having an answer to act on. FastQC flags on lane 3 live in the same thread where you're planning the re-run. A DESeq2 gene list sits beside the target rationale and literature evidence you're already building.

For R&D leaders running RNA-seq as part of target validation, biomarker discovery, or mechanism work, that means faster QC, reproducible differential expression, and outputs that land where the science actually happens, without adding a new tool for anyone to learn.

Try it on your next dataset

Have FASTQ files from a recent run? Attach them as a Bioinformatics File and ask EMET to run FastQC, or walk through a differential expression analysis from raw reads to a gene list.