Know our pathogenicity predictors
To assess the functional significance of missense variants, we have calculated the best predictor for more than 50 proteins involved in neonatal diseases. Predict your variant or check our examples to see the result
Prediction of the functional significance of missense variants of more than 50 proteins involved in neonatal diseases
Disclaimer This resource is uniquely intended for research purposes. The authors are not responsible for neither its use nor misuse. The data provided are not intended as advice of any kind. The authors have worked with care in the development of this server, but assume no liability or responsibility for any error, weakness, incompleteness or temporariness of the resource and of the data provided.
How to interpret a prediction?
The variants are classified as pathogenic or neutral by a supervised machine learning algorithm
The label depends in the numeric score. Variants above 0.5 are pathogenic while variants low 0.5 are neutral
Beyond the label, a variant is characterized with a numerical score of its pathogenicity. We provide you with this score, so you can have a more precise sense of the consequence of the variant
This score has a continuous range between 0 to 1, being 0 a neutral and 1 a pathogenic variant. The threshold between pathogenic and neutral variant is at 0.5
We have evaluated our predictors with rigorous quality controls and compared to the state of the art predictors such as PolyPhen2, PON-P2, SIFT or CADD
To evaluate our predictor, we perform a leave-one-out cross-validation. From it, we derived several performance metrics such as sensitivity, specificity, accuracy, AUC and MCC