Pricing
Start with a single target, scale to a multi-target campaign, or unlock unlimited access for your discovery program. Discovery and Research tiers are customisable — annual prepay, multi-seat and multi-year terms negotiable for serious buyers.
Explorer
Single-target pilot
- ✓Top-3 ranked pockets (geometric ranker)
- ✓Conformational ensemble + 3D viewer
- ✓Pre-pilot MSA tractability screen
- ✓PDF report + downloadable bundle
- ✓Per-pocket geometric metadata + residue lists
- ✓14-day data retention
- ✓Email support
For initial target assessment
Discovery
Multi-target campaign · up to 5 targets
- ✓Everything in Explorer
- ✓Up to 5 targets across the 3-month window
- ✓2 hours methodology consultation
- ✓Shared interpretation calls
- ✓Pre-pilot screen on every target
- ✓30-day data retention
- ✓Email + scheduled review support
For active short-term programs
Research
From €50,000 with annual prepay
Unlimited targets
- ✓Everything in Discovery
- ✓Unlimited targets
- ✓API access
- ✓Opt-in MSA-driven prediction (where the screen indicates)
- ✓Priority queue
- ✓90-day data retention
- ✓Dedicated support
For active RNA-targeted discovery programs
Enterprise
Everything in Research plus
- ✓Dedicated compute instance
- ✓Zero data retention
- ✓On-premise option
- ✓SLA
- ✓Custom support
info@rnafold.com
Frequently asked questions
What does v0.2 actually return?
For each target sequence: a predicted 3D structure, a 5-frame conformational ensemble, and a top-3 ranked list of candidate pockets. Pockets are detected with RNA-tuned cavity parameters (consistent with the published fpocketR approach, Veenbaas et al. PNAS 2025) and ranked by structural persistence across the ensemble weighted by binding-residue stability. Druggability assessment itself is left to your medicinal chemistry workflow; we provide the geometric metadata and conformational stability metrics as inputs to that workflow.
How does it perform on benchmark targets?
On a 7-target cleft-binder benchmark with deposited co-crystal structures, the locked v0.2 pipeline (ensemble + cross-frame geometric ranker) recovers the binding-site cluster at rank-1 in 3 of 7 cases at strict@1 (overlap ≥ 50%) and 6 of 7 at near@1 (overlap ≥ 30%). RNA-tuned cavity detection alone, without the ensemble + ranker on top, recovers 0 of 7 at strict@1 and 2 of 7 at near@1 — equivalent whether using our parameters or the published fpocketR parameters. The ensemble + ranker is the load-bearing v0.2 contribution. Full per-target comparison table on /methodology; live demos at /druggability.
When is MSA mode used?
A pre-pilot screen estimates whether your target's MSA carries enough evolutionary diversity for MSA-driven prediction to help. Targets with at least one homolog at <77% identity (or a non-trivial fraction of homologs in the 70-80% identity band) get offered MSA mode; otherwise we default to single-sequence prediction. The criterion is empirical, calibrated on our benchmark, and refined as more targets accumulate.
Are my sequences private?
Explorer: input sequences may be used to refine our pre-pilot screen calibration and benchmark composition. Discovery, Research and Enterprise: your sequences are never used for training or calibration and are deleted after the retention period. All processing on EU infrastructure.
What RNA lengths are supported?
Up to 500 nucleotides on Explorer, Discovery and Research plans. Longer sequences available on Enterprise.
Can the listed prices be customised?
Yes — Discovery and Research are negotiable for serious buyers. Annual prepay, multi-seat licensing, multi-year terms and campaign-volume discounts are all on the table. The list prices are the starting point; email info@rnafold.com to discuss what fits your program.
What about groove binders?
Out of scope for v0.2. Our pipeline detects cleft-shaped binding pockets; groove-binding modes (extended sequence-recognition interfaces, e.g. tRNA-recognising riboswitches or some aminoglycoside-class targets) may be missed. Contact us if your target's binding mode is groove-mediated — we'll be transparent about whether we expect to help.