Sentence examples for dti prediction from inspiring English sources

As stated in Bleakley and Yamanishi (2009) and van Laarhoven et al. (2011), when evaluating the DTI prediction results, in which there are usually few positive DTIs, PR curves provide greater biological significance and are considered a better quality measure than ROC curves.

Although previous network-based approaches have achieved promising results for DTI prediction and drug repositioning, few of them are specifically designed for integrating and predicting different types of DTIs on a multidimensional network.

Except for the predictive model, similarity measuring is another crucial factor in DTI prediction because similar drugs tend to interact with similar targets [ 11].

The parameter D ij can be also learned using the CD algorithm: (13) We chose a conditional RBM to perform DTI prediction on a multidimensional network that encodes different types of DTIs.

Two other customary ways would be to count a predicted DTI for a drug as correct, if it is top-ranked after removing known DTIs for that drug as used in [22], or to count a DTI prediction as correct if it is within the 'top 1'% (5 or 10%%) of all predictions for the drug.

Despite these positive aspects, current rich information about types of DTIs (Günther et al., 2008; Kuhn et al., 2012) has not been well exploited for DTI prediction, and how to incorporate such information into a multidimensional network to predict different types of DTIs still remains an open question.

Moreover, on the same datasets, the single top ranked DTI predictions by DASPfind are correct on average in 49.22 % of cases when predicting any of the known DTIs for a single drug, assuming there are no known DTIs for the drug.

Our method for this dataset generates 210 predictions as the 'top 1' DTI predictions (one for each drug).

DASPfind outperformed the other methods we evaluated with respect to retrieving known interaction as the 'top 1' DTI predictions using all other datasets.

For the same dataset, in LOOCV, 62.74 % of the known interactions were retrieved among the 'top 5' DTI predictions by DASPfind, while among the 'top 5' DTI predictions by NRWRH, HGBI and DT-Hybrid only 12.89, 12.41, and 10.95 %, respectively, were the known ones.

Out of these 210 DTI predictions, 91 are already known (the dataset we used has not been updated since 2008), while the remaining predicted DTIs are not present in that dataset.