No account to create, no minimum order, and a fixed price agreed before anything starts. This is the whole process, including the parts most vendors leave vague.
Sample count, species, platform, and the comparisons you actually want. Two minutes on the quote form is enough.
Scope, price and a delivery date, normally within one working day. If the data turns out to be different from the description, we re-quote before starting rather than after.
Count matrices where possible, your own cloud link or ours. NDA first if you want one. Nothing starts before you approve the quote.
We run quality control before the rest and report what we find — including problems. A sample that should be dropped is dropped from the scope and the invoice.
You see clustering and annotation before we build the final figures, so a wrong assumption costs an hour rather than a week.
Figures, results tables, the processed object, the methods paragraph, and the code with its environment. Then six months of reviewer-driven revisions on that analysis.
Count matrices (Cell Ranger filtered_feature_bc_matrix or .h5), h5ad, Seurat .rds, plain csv/tsv matrices, Space Ranger output for spatial, and raw FASTQ when re-alignment is genuinely needed. If you are not sure, send the folder listing and we will tell you which parts matter.
A standard single-cell project is typically five to eight working days from receiving usable data. Priority turnaround is possible. Most delays are not computation — they are missing sample metadata and undefined comparisons, which is why we ask for both up front.
We tell you before doing the rest of the work, with the evidence. A failed sample can be dropped from the scope and from the invoice; we would rather lose part of a project than deliver an analysis built on a sample that should not be in it.
Yes, on request, before any data moves.
It is deleted after delivery on your instruction, and by default within 90 days. It is never reused for benchmarking, marketing examples, model training, or anything else. Unpublished data is the most valuable thing a lab owns.
You do — results, figures, and the code we write for your project. We keep no rights over your data or your findings.
That is the point of shipping the code and the environment. Every project comes with the script, the parameter set and the package versions, so any figure can be regenerated after we are gone.
Reviewer-driven re-runs of the delivered analysis are included for six months. New analyses beyond the original scope are quoted separately, usually as a small add-on.
No. This is analysis only, which is why there is no minimum order and no obligation to buy anything else. If you need sequencing we are happy to tell you what to ask a provider for.
Three complete analyses of public datasets are published on this site, including one where an automated classifier was run blind against our manual annotation and the two were compared cluster by cluster. You can also run the free QC report on your own data first.
We will tell you what the data supports and what it does not, and we flag results we think are fragile. The scientific interpretation stays with you — that is your paper.
→ Get a fixed quote · run the free QC report first · see a complete sample deliverable