The intelligence layer
for your
single-cell omics data

"The recurring problem I faced was getting inconsistent annotations from different team members, then being stuck trying to decide between them without clear criteria. CyteType's multi-reviewer system and pathway enrichment let me evaluate the biology myself. It's not perfect for rare populations, but it gets us most of the way there much faster than our previous workflow."

Principal Investigator — Karolinska Institutet

"This was my first time analyzing scRNA-seq independently and the computational aspects were challenging. The cluster chat feature was really helpful - I could ask why certain plasma cells showed unexpected markers and get explanations grounded in the actual expression data. The speed was remarkable, what I'd budgeted three weeks for took an afternoon. My supervisor was impressed with how quickly I could defend the annotation choices."

PhD Student — Università degli Studi di Salerno

"We support approximately 60 projects per year across different tissue types. Building and maintaining reference datasets for each context was becoming unsustainable. CyteType eliminates that requirement entirely, and the accuracy is comparable to our carefully curated references. The main advantage for us is consistency - every project now starts from the same standardised baseline. We still validate everything, but the time savings are significant."

Senior Computational Biologist, Core Facility — Institut Pasteur

"The functional state resolution was the standout feature for me. Instead of simply labeling cells as macrophages, we got detailed breakdowns like 'M2-polarized macrophages with active lipid metabolism and ECM remodeling', with specific gene sets supporting each functional program. This granularity completely changed our validation plan - we knew exactly which biological processes to target in follow-up perturbation studies."

Researcher, Computational Oncology — Memorial Sloan Kettering Cancer Center

"I've evaluated several automated annotation tools. CyteType's strength is the reasoning transparency - you can examine exactly why each label was chosen and what the sources of uncertainty are. My one concern is potential bias from the LLM's training data. If the model has encountered papers on a specific tissue type, is it discovering patterns de novo or recalling literature? We independently verify key populations, but it's worth being aware of. The heterogeneity scores have been useful for identifying clusters that need subclustering, and for first-pass annotation it performs well."

Bioinformatician — Helmholtz Zentrum München

"Our patient samples contained many transitional microglial states that were difficult to classify. What proved most valuable was how CyteType decomposes cell states into discrete functional programmes - clearly separating which genes support activation versus antigen presentation versus proliferation. This gave us clear hypotheses about what these cells were actually doing in the disease context, rather than just arguing over terminology."

Postdoc — Ludwig-Maximilians-Universität München

"Annotation is necessary but time-consuming work that doesn't directly generate publications, yet incorrect annotations can derail a manuscript. CyteType handles the majority of straightforward clusters, allowing my team to focus on biologically interesting populations. The CytetypeR package integrated smoothly with our existing Seurat pipeline - just a few lines of code. No infrastructure changes required. For differentiated cell types, it's proven quite reliable."

Principal Investigator, Developmental Biology — University College London

"The multi-reviewer assessment caught something we'd initially missed. One reviewer flagged unexpectedly high HLA-DR expression in what we'd labeled as NK cells, which prompted us to reclassify them as activated CD8+ T cells. The literature citations were also valuable when responding to reviewer questions about our monocyte subtype assignments."

Researcher, Infectious Disease — University of Innsbruck

"As someone still learning the field, the detailed marker-level explanations were really educational. Rather than just accepting a cell type label, I could see which specific genes supported each call and understand their biological relevance. This made me much more confident discussing the data in journal clubs and lab meetings."

PhD Student — Medical School of Hannover

"The functional state annotations go well beyond basic cell type labels. For our tumour microenvironment study, seeing the specific metabolic and signalling programmes active in each immune population helped us understand the biology mechanistically. This directly informed better experimental design."

Bioinformatician — University of Glasgow

"We advise multiple research groups, and maintaining annotation consistency across projects has been challenging. Building and maintaining reference atlases for every tissue type was consuming significant resources. CyteType performs comparably to our carefully curated references without requiring that upfront investment. The Cell Ontology mapping makes meta-analyses much more feasible."

Senior Computational Biologist — University of Cambridge

"In translational research, you need annotations that can withstand scrutiny. The complete evidence trail - markers, citations, reviewer assessments - provides documentation we haven't had access to before. We've used it for immune monitoring studies, and the clinical team appreciated being able to trace the reasoning behind each classification. It still requires domain expertise to interpret appropriately, but it provides a solid starting point."

Clinician-Scientist — Mass General Brigham

"We study disease processes across multiple tissue types, and maintaining cross-tissue annotation consistency has been difficult. CyteType's context-aware approach has helped us compare related cell populations in different organs more systematically. The tool doesn't replace expert biological knowledge - we've corrected some edge cases - but it dramatically accelerates the process and improves reproducibility across our lab."

Principal Investigator — University of Florida