Open Zainab-ik opened 5 months ago
GemmaTuron
commented
5 months ago Hi @Zainab-ik
We need to look if the data is open: In the present study, data from six different studies,41−46 comprising 4119 drugs, were combined to construct a training set for predicting the fraction bound (f b) in plasma.
Zainab-ik
commented
5 months ago Hi @GemmaTuron
For this model, there's no single data source or combined resources. All the data sources are scattered. In ref 41 - supplementary information for the dataset - https://static-content.springer.com/esm/art%3A10.1007%2Fs11095-013-1023-6/MediaObjects/11095_2013_1023_MOESM4_ESM.xlsx in ref 42 - there's a directory in the repository for the data - https://github.com/rtv2016/PPB/tree/master/data in ref 43 - the dataset is not opened. in ref 44 - dataset shown in the publication - https://www.sciencedirect.com/science/article/pii/S0006295202010742?via%3Dihub#TBL1 in ref 45 - dataset is not available in ref 46 - dataset is in the supporting information - https://pubs.acs.org/doi/abs/10.1021/acs.molpharmaceut.8b00785
I've added the available datasets to the annotation
GemmaTuron
commented
5 months ago But if they are available we should cite the six data sources they use
Summary !!!
BioModel Name: Jiménez-Luna2021 - Coloring molecules for plasma protein binding prediction
BioModel Tag: Machine learning model, Ersilia
Metadata Comments: The endpoints for this model is plasma protein binding prediction.
Contributor:
Curation status : manually curated
Annotation file: here