A: ProtoPRED can produce four different types of output: the results page always includes a summary table with all the predictions performed and an additional model-specific table with additional details for each different model, as well as an option to download an Excel spreadsheet (with more information such as the values in the internal model units). Additionally, you can download standardised PDF reports (this could have an additional cast) such as QPRFs of all the QSAR predictions (except for ProtoNANO) and specific combined reports for ProtoICH and GenoITS.
A: ProtoPRED uses calculated molecular descriptors to characterise your substance. In some cases, these descriptors cannot be calculated due to numerical issues or incompatibilities between the structure and the descriptors. When this occurs, ProtoPRED uses imputation to estimate the values from our database. However, this is only permitted for a limited number of descriptors. If an excessive number of values require imputation, the prediction will not be provided because it will not be considered a reliable predcition. In such cases, the results will be included in the Excel and HTML outputs, but a PDF report cannot be downloaded as no prediction is available.
A: QSAR models are trained using a database of molecules and experimental values, so their validity is restricted to the chemical space defined by this dataset. ProtoPRED uses four different techniques to evaluate this chemical space and to evaluate if your target molecule is within it. A prediction is always provided, but those outside the applicability domain are considered less reliable.
A: Assessing the reliability of a calculation implies factors such as the overall predictivity of the model, the relationship of the target molecule with the training set, the behavior of model with similar substances and the adequacy of the input. To help the user, ProtoPRED provides a combined reliability score using different parameters (list of parameters and individual values are in the QPRF). Note that this score is presented as a guidance and does not substitute a tailored expert assessment considering the details of each factor included in the QPRF.
A: No. The information on analogues is provided to assess the prediction. Analogues may or may not be part of the model's training set but do not contribute directly to the prediction. However, they do influence the reliability score of the prediction.
A: You can provide content for various sections of the report, such as those regarding your data, the identity of the substance, and additional comments in different sections. However, it is currently not possible to modify the existing content directly. ProtoPRED offers an automatic QPRF tool to assist users, but the QPRF is submitted by the user, allowing you to use the information to create your own version of the document.