This motivates us to employ the semi-supervised method that can utilize the information from the unlabeled samples to predict drug-protein interaction.
Furthermore, a prediction method that can utilize low-resolution models is advantageous because high-resolution structures are available for only a relatively small fraction of proteins.
Because many time series gene expression data sets are now available in the public domain [ 14- 18], it is desirable to develop a module inference method that can utilize the dynamic nature of time series gene expression data.
Furthermore, as these methods can utilize random utility theory (RUT), they are more consistent with economic [ 48] and psychological [ 49] theories of choice.
We named this centrality value the "node-limited betweenness centrality (nlBC)", and this method can utilize the sum of the weight values of both ends of each shortest path (see "Methods" for details).
The method can utilize multiple targets (in this paper, Si and C) simultaneously in order to deposit complex compositional coatings.
The combination of appearance model and traditional machine learning techniques consumes target tracking as a binary classification problem [6,7], and this method can utilize background information effectively and thus can improve the effectiveness of tracking.
For example, our method can utilize the following path over the network: drug1 (start) → protein1 → drug2 → protein2 (end), which maps to the following example path in the bottom-right side of Fig. 1: D00316 → hsa5915 → D00094 → hsa6096.
However, our method can utilize the information of the moments that are from second moment to high moment when it is necessary according to the sample size and data.
The GREML method can utilize data on very distantly related individuals, individuals that are typically considered "unrelated" in traditional GWAS.
In contrast to NBH-based prediction of numerical potency values, the conditional probability method can utilize approximate potency measurements (e.g., primary screening data) leading to a binary classification of inactive vs. active data set compounds and ensuing prediction of a probability of activity for VCs.
