Privacy policy

Effective date: 17 November 2026

rnafold.com is operated by JBR Technologies Ltd. from Malta, EU. We comply with GDPR and applicable EU data protection regulations.

Data we collect

When you submit a job, we collect: the sequences you supply (RNA, DNA or protein), any structure files you upload, your email address, and usage metadata (timestamps, IP addresses for rate limiting).

If you use the API, connect an AI assistant, or use projects, we also store the following.

  • API keys. You can create keys on the Analyses page. We keep only a one-way hash of each key, never the key itself, together with the label you gave it and when it was created and last used. A key acts as your account and uses your allowance. It cannot create or revoke keys.
  • Connected apps. When you connect an app such as claude.ai by signing in, we record the app's name and the address it sends you back to, and we issue it a key that lasts one hour and renews itself with a refresh token valid for 30 days. A connected app appears in your API keys list, and revoking it there ends the connection at once.
  • Projects and notes. A project holds a name, an optional description, notes written by you or by your assistant, and references to your own analyses, runs and groups. Projects and their notes are kept until you delete them. They do not expire after 14 days, so please do not put anything in a note that you would not want kept.
  • Groups, pipeline runs and templates. The sequences and settings you save for reuse.

How we use your data

  • We do not train models. Sequences and structures you submit are used only to produce your own results. They are never used to train or fine-tune any model, on any tier. The third-party structure prediction models we run are used for inference only and are never fine-tuned.
  • Email addresses are used to send job completion notifications and account messages, and, occasionally, to write to you about the service you use: changes to it, and to ask for your feedback on how it works for you. We do not send advertising for other companies and do not share your address for marketing. You can ask us to stop these messages at any time by replying to one or writing to info@rnafold.com; job completion notifications and account messages continue. Our legal basis for them is our legitimate interest in improving a service you use (GDPR Art. 6(1)(f)), and you have the right to object.
  • Nothing you submit is shared with other users, and no data is sold.

Where your data is processed

Prediction runs on EU infrastructure. The GPU instance is located in Germany, and the job queue runs on that same machine rather than a hosted third-party queue, so submitted sequences are not passed to an external queue service. Downloadable result artefacts are stored in a Cloudflare R2 bucket created under EU jurisdiction.

Requests to our API are carried by Cloudflare's network, where the encrypted connection is terminated before the request reaches our server in Germany. The computation and the storage of what you submit are in the EU as described above.

Data retention

  • Free tier: jobs, submitted sequences and results are deleted automatically 14 days after submission.
  • Paid tiers: retention is agreed per engagement, including deletion immediately after delivery.
  • Groups, pipeline runs and templates: kept for 30 days on the free tier and 365 days on paid tiers, or until you delete them.
  • Projects and notes: kept until you delete them. Deleting a project removes the project and its notes, not the analyses it refers to, which expire on their own schedule.
  • API keys and connected apps: until you revoke them. Revoking takes effect immediately.
  • Usage-limit counts: when an account reaches a usage limit, we keep a count of how often, against the account identifier, for 90 days. It holds no sequences or structures.

Records that a job ran — timestamp, account identifier, tier, molecule type, sequence length and duration — are retained beyond this period for service metrics. They do not contain your sequences or structures.

Data security

Sequences are encrypted in transit (TLS), and stored result artefacts are encrypted at rest by our storage provider. Access is restricted to authorised personnel only.

Logs

Our servers keep operational logs for security, rate limiting and troubleshooting: the time, the request path, the response status and the IP address. For the MCP server a log line also carries a short hash of the key and, for a request we refuse, the kind of request and the protocol version. These logs do not contain submitted sequences, the contents of files or requests, or API keys, and they stay on the same machine in Germany.

Your rights

Under GDPR you have the right to access, rectify, or delete your personal data. To exercise these rights, contact info@rnafold.com. Deletion requests are honoured within 30 days.

AI assistants

We do not send your data to any AI assistant. An assistant you connect is a separate service that you choose: what you type to it, including a sequence, is handled under that provider's terms and privacy policy, not ours. If a sequence must not leave EU infrastructure, do not paste it into an assistant hosted elsewhere.

Tools that answer in seconds, such as secondary-structure prediction, primer design and codon recoding, compute their answer in memory and return it. What you send to them is not stored and is not written to our logs. Folds and other analyses that run on the GPU are stored as described under data retention.

Third parties

We use Vast.ai (EU GPU compute), Cloudflare (CDN, R2 storage, and the network that carries requests to our API), Clerk (authentication) and Resend (email). No data is sold to third parties.