Frequently Asked Questions
What does the Aptamer Designer actually do?
Given a target amino-acid sequence, the service iterates over candidate DNA aptamer variants (a variable region flanked by fixed prefix and suffix), scores each with a machine-learning interaction model, folds it with ViennaRNA using DNA thermodynamics, and returns the top 25 highest-scoring candidates.
What does the interaction probability score mean?
It is the output of an XGBoost classifier trained on known aptamer–protein pairs. Values are in [0, 1]; higher values indicate a candidate the model considers more likely to bind the target.
What is MFE and why can it be positive?
MFE is the Minimum Free Energy of the predicted secondary structure in kcal/mol. More-negative values mean a more stable fold. Values near zero (or positive) mean the sequence is unlikely to fold into a stable structure under the DNA thermodynamic model.
Why does my job take a while?
Each candidate sequence is filtered by GC content and repeat rules, then scored by the ML model, then folded. For large search spaces (e.g. variant length 12+ with many target sequences) this adds up. Progress and estimated remaining time update while the job runs.
How long are results kept?
Result URLs stay valid for 90 days from job submission. Bookmark the result URL if you need to return to a specific job.
Can I submit a custom prefix and suffix?
Yes. Toggle the custom-sequences option in the Aptamer Designer form and enter DNA prefix / suffix strings (A, T, G, C only). The variant region is generated between them.