Common questions about CyteType. For confidence thresholds and edge-case review habits, see Limits and interpretation.
General
❓ What is CyteType?
CyteType is a structured AI workflow for automated annotation of single-cell RNA-seq data. Specialized steps evaluate marker expression, reference similarity, ontology structure, and literature context in a fixed order, and their outputs are merged into a final annotation with confidence scores and traceable reasoning. The full method is described in the CyteType preprint (Ahuja G et al., bioRxiv 2025).
❓ How is CyteType different from existing tools?
Traditional approaches depend on a single reference or a set of marker genes. CyteType instead integrates several biological signals through a structured agent workflow, leading to higher robustness in rare, transitional, or disease-associated cell populations.
❓ Why is the structured workflow important?
The performance gains originate from the workflow rather than the LLM tier. Each agent contributes a different biological perspective, and a reconciliation step produces a stable, evidence-supported annotation.
❓ Who developed CyteType?
CyteType was created by Nygen Analytics, a research-focused biotech company in Sweden working on AI systems for single-cell omics.
❓ Which programming environments are supported?
CyteType is available in Python for AnnData/Scanpy workflows and in R (CyteTypeR) for Seurat.
❓ How do I install CyteType?
Python: pip install cytetype (Python ≥ 3.12). R: devtools::install_github("NygenAnalytics/CyteTypeR").
❓ Which input formats can I use?
AnnData and Seurat objects. Standard Scanpy or Seurat preprocessing workflows do not require reformatting.
❓ What resources are needed to run CyteType?
CyteType requires internet connectivity for annotation and artifact upload. Local preprocessing time, upload time, and memory use depend on the size and storage layout of the analysis object. For large AnnData objects, use backed mode so the client can process the matrix without loading the complete object into memory.
Setup and data
❓ Does CyteType compute marker genes?
CyteType expects user-provided cluster marker genes as priors. Marker selection is often study-specific, so the tool does not override user-defined markers. CyteType supplements these priors by analysing pseudobulked cluster profiles to identify additional genes outside the provided list.
❓ Why do I need to compute markers beforehand?
Marker gene computation is left to the user to maintain compatibility with different preprocessing workflows and clustering strategies. CyteType uses these markers as a structured informational anchor during reasoning.
❓ Why does CyteType pseudobulk clusters?
Pseudobulk profiles allow CyteType to evaluate genes that are not included in user-provided marker lists. This expands the evidence base for annotation and improves accuracy in transitional or poorly characterised populations.
❓ Which species can be annotated?
Human and mouse datasets are fully supported. Other species including rat and zebrafish can be analysed through ortholog mapping, subject to gene homology quality.
❓ Are tissue-specific reference sets included?
Yes. Reference sets optimised for PBMC, brain, liver, kidney, lung, and pancreas are provided. Custom references can also be supplied.
❓ Does CyteType annotate disease samples?
Yes. Because the method does not rely solely on healthy-tissue references, it performs well on tumour samples, perturbed systems, and transitional states.
Analysis
❓ How accurate is CyteType?
CyteType was benchmarked across 20 datasets and 977 clusters. Using the CyteOnto semantic similarity metric against author-assigned labels, CyteType achieved up to 3.8-fold improvement over GPTCellType, 2.68-fold over CellTypist, and 1.01-fold over SingleR. Forty-one percent of clusters received enhanced functional annotation and twenty-nine percent received refined subtype resolution relative to author labels.
❓ How long does annotation take?
Typical runtime is two to three minutes per cluster. A dataset of about fifteen clusters usually completes in thirty to forty-five minutes. The system supports high concurrency for large studies.
❓ Are annotations assigned at the cell or cluster level?
CyteType assigns annotations at the cluster level and propagates them to individual cells. This balances computational efficiency with biological resolution.
❓ Why are repeated runs similar but not identical?
CyteType uses a structured, server-managed workflow that evaluates biological evidence in a defined sequence. The underlying models can still produce variation in wording, reasoning, confidence, or the final label. Treat agreement across runs as supporting evidence rather than a guarantee, and validate important calls against markers and study context. query.json records the submitted request, but it does not identify the complete server release, provider route, or uploaded artifact state.
❓ How should confidence scores be interpreted?
Confidence values are provided for cell type, subtype, and activation state. Scores above 0.8 generally indicate high reliability. Lower scores mark ambiguous or poorly supported populations that warrant manual review. See Confidence and heterogeneity QC for how to act on mixed or low-confidence clusters.
❓ What does CyteType return?
CyteType adds annotation metadata to the input object, including labels, ontology terms, confidence scores, and evidence summaries. A standalone HTML report provides interactive reasoning, citations, and interpretation tools.
❓ How does Cell Ontology linking work?
CyteType maps annotations to official Cell Ontology (CL) terms. When subtype precision is limited, it proposes the closest parent term and includes ranked alternatives.
❓ How is literature evidence generated?
The Literature and Context agent uses LLM-assisted retrieval to identify relevant publications and summarise supporting evidence. Citations are included for independent verification.
❓ What does anomaly detection identify?
Clusters with unexpected marker signatures are flagged. These can represent doublets, low-quality populations, or potentially novel biological states.
❓ Does the 1000-cell UMAP display affect annotation?
No. UMAP subsampling is purely for rendering. All cells contribute to cluster-level pseudobulk profiles, and annotations are based on full data.
❓ Can CyteType restrict reasoning to specific tissues or lineages?
CyteType does not enforce strict tissue constraints. Reasoning is influenced by the study_context provided by the user, but agents are free to consider the full biological landscape. This avoids prematurely excluding plausible alternatives.
Privacy and security
For the complete data flow and control description, see Security and privacy.
❓ What data does CyteType upload?
The official clients submit study context, marker genes, cluster-level expression summaries, metadata summaries, and sampled visualization coordinates. They also upload a normalized expression matrix with feature metadata and raw counts when available, plus the complete cell observation metadata table. They do not upload FASTQ files, BAM files, or the original AnnData or Seurat object as a single file. See Data handled in a standard CyteType run.
❓ Is scientific data used for training or benchmarking?
No. Uploaded scientific data is not used to train or benchmark models. CyteType does collect operational telemetry to run, secure, support, and improve the service, including service-health, performance, error, and product-interaction information. See How CyteType uses scientific data.
❓ How does CyteType protect data and report access?
CyteType uses HTTPS with TLS for data in transit and provider-managed encryption for stored data. Passwordless email sign-in, API tokens, secure session cookies, signed report-access tokens, server-side authorization checks, rate limiting, and monitoring protect access. See How CyteType protects access.
❓ Who can access my report?
New reports are private by default. The owner can share a private report with specific users, make it visible to the organization, or make it public. Organization administrators do not gain access to another user's private report through the administrator role alone. Authorized Nygen personnel may access service data when required to operate, secure, support, or troubleshoot CyteType. See Access and sharing controls.
❓ Are third-party model providers involved?
Yes. Model providers process the focused scientific context needed for their workflow step. This can include study context, markers, cluster-level expression percentages, metadata summaries, and annotation hypotheses. Complete vars.h5 and obs.duckdb artifact files are not sent to model providers. Managed provider routes can vary by workflow step and service availability. See Models, infrastructure, and scientific services.
❓ Can I use my organization's model provider?
Yes. CyteType supports customer-supplied API configurations, including OpenAI API-compatible services and AWS Bedrock. A ChatGPT Enterprise workspace is separate from OpenAI API access. A customer-owned provider changes the inference account, but the CyteType backend still receives the documented inputs and hosts the report. See Use your organization's model provider.
❓ Can I delete my report?
Yes. Report owners can delete reports in the CyteType Dashboard. Deletion removes the report from normal dashboard views, revokes standard report access, and removes explicit user shares. See Deleting reports.
❓ Can I use CyteType with confidential human data?
CyteType does not detect or remove direct identifiers or protected health information. Work from a reviewed copy of the analysis object, remove identifiers and unnecessary observation columns, and complete the required institutional review before uploading confidential data. The current official clients do not provide a summary-only mode that disables artifact upload. See Plan for sensitive or regulated work.
❓ Is a Data Processing Agreement available?
Yes, for customers using CyteType under a commercial license. Contact contact@nygen.io to request a DPA, complete a security questionnaire, or discuss customer-specific requirements. See Complete your security review.
Pricing and access
❓ Is CyteType free for academic use?
Yes. Academic and non-commercial use is free. Personal accounts can submit up to 10 annotation runs in a rolling 24-hour window, 50 in 7 days, and 100 in 30 days. Failed jobs do not count toward these limits. CyteType is licensed under CC BY-NC-SA 4.0.
❓ How can I exceed the daily limit?
Organization and commercial licenses can provide shared or customer-specific quotas. Contact contact@nygen.io to discuss expected use. Supplying a model-provider account changes the inference configuration but does not remove CyteType account limits.
❓ How is commercial use licensed?
Licensing includes an annual fee and a per-cluster annotation cost. Nygen offers Discovery, Enterprise, and Partnership tiers. Contact contact@nygen.io for details.
❓ What options are available for institutional use?
CyteType supports organization workspaces, shared quotas, customer-owned model-provider accounts, AWS Bedrock, and DPAs for commercial licenses. Contact contact@nygen.io before uploading data if your institution requires a specific provider, access procedure, or deployment boundary.
Support
❓ Does CyteType fit into existing Scanpy and Seurat workflows?
Yes. CyteType writes outputs directly into object metadata (adata.obs/obj@meta.data). Existing downstream pipelines remain unchanged.
❓ What preprocessing is required?
Standard QC, normalisation, and clustering. Apply batch correction (such as Harmony or Scanorama) when working with multi-batch datasets before submission.
❓ Can I use custom marker lists or references?
Yes. Custom marker genes and reference datasets can be supplied for specialised tissues or perturbation studies.
❓ Does CyteType support multimodal data?
CyteType currently expects transcriptomic input. Support for multimodal and CITE-seq data is under evaluation.
❓ How do I add custom metadata to my report?
Pass metadata as a dict (Python) or named list (R) to the run step. Keys appear as headers in the report. Values that look like URLs are rendered as clickable links, useful for linking GEO accessions, DOIs, or internal data portals.
❓ What if I disagree with an annotation?
The HTML report allows manual override, re-annotation requests, and reasoning queries via the Cluster Copilot. All changes are logged.
❓ How does CyteType address ambiguous populations?
The Reviewer agent highlights ambiguity and provides ranked alternative hypotheses with supporting evidence. Low confidence scores indicate clusters that warrant closer inspection.
❓ My job is taking a long time. Is it stuck?
Jobs continue running on the server even after your local session disconnects. Use annotator.get_results() (Python) or GetResults(obj) (R) to retrieve results after reconnecting. The report URL printed during submission updates live; check it to see per-cluster progress.
❓ The artifact upload is failing. What should I do?
require_artifacts=False does not skip artifact building or upload. It allows the run to continue if one or more artifacts fail, while successfully built artifacts are still uploaded. Review the reported upload error before using this fallback because the interactive report can lose expression lookups or metadata filtering. For large datasets, load AnnData in backed mode: sc.read_h5ad("file.h5ad", backed="r").
❓ Does CyteType detect doublets or low-quality cells?
Potential doublets and low-quality populations are flagged through confidence scoring. Removal should occur during preprocessing before submission.
❓ Is spatial transcriptomics supported?
Spatial support is in development and will be released after validation.
❓ Can I use a local model with Ollama?
Yes. Expose your local Ollama instance using ngrok, then pass the public URL as baseUrl with provider="openai". The model must support tool calling. This changes the model-inference endpoint, but the CyteType backend still receives the documented inputs and hosts the report. See the Ollama integration guide for step-by-step instructions.
❓ How often are databases and references updated?
Cell Ontology updates are synced quarterly. Literature context is refreshed regularly. Workflow improvements are deployed automatically.
❓ How do I cite CyteType in my publication?
Please cite the bioRxiv preprint: Ahuja G et al., Multi-agent AI enables evidence-based cell annotation in single-cell transcriptomics, bioRxiv 2025. doi:10.1101/2025.11.06.686964
❓ Where can I get support?
GitHub: github.com/NygenAnalytics/CyteType, Discord: discord.gg/V6QFM4AN, Support: support@nygen.io
❓ Is there a tutorial or interactive notebook?
Yes. A Google Colab tutorial is available for an interactive end-to-end walkthrough: Open in Colab
Learn more
