Limits...
Supervised prediction of drug-target interactions using bipartite local models.

Bleakley K, Yamanishi Y - Bioinformatics (2009)

Bottom Line: We demonstrate the excellent performance of the proposed method in the prediction of four classes of drug-target interaction networks involving enzymes, ion channels, G protein-coupled receptors (GPCRs) and nuclear receptors in human.This enables us to suggest a number of new potential drug-target interactions.An implementation of the proposed algorithm is available upon request from the authors.

View Article: PubMed Central - PubMed

Affiliation: Mines ParisTech, Centre for Computational Biology, Fontainebleau, France. kevbleakley@gmail.com

ABSTRACT

Motivation: In silico prediction of drug-target interactions from heterogeneous biological data is critical in the search for drugs for known diseases. This problem is currently being attacked from many different points of view, a strong indication of its current importance. Precisely, being able to predict new drug-target interactions with both high precision and accuracy is the holy grail, a fundamental requirement for in silico methods to be useful in a biological setting. This, however, remains extremely challenging due to, amongst other things, the rarity of known drug-target interactions.

Results: We propose a novel supervised inference method to predict unknown drug-target interactions, represented as a bipartite graph. We use this method, known as bipartite local models to first predict target proteins of a given drug, then to predict drugs targeting a given protein. This gives two independent predictions for each putative drug-target interaction, which we show can be combined to give a definitive prediction for each interaction. We demonstrate the excellent performance of the proposed method in the prediction of four classes of drug-target interaction networks involving enzymes, ion channels, G protein-coupled receptors (GPCRs) and nuclear receptors in human. This enables us to suggest a number of new potential drug-target interactions.

Availability: An implementation of the proposed algorithm is available upon request from the authors. Datasets and all prediction results are available at http://cbio.ensmp.fr/~yyamanishi/bipartitelocal/.

Show MeSH

Related in: MedlinePlus

PR curves for predicted drug–target interactions using BLMs on four benchmark datasets: (a) enzyme, (b) ion channel, (c) GPCR and (d) nuclear receptor. The solid line is for leave-one-out on potential drugs (row 2 of Tables 1–4), the dashed line for leave-one-out on potential target proteins (row 5 of Tables 1–4) and the dotted line for aggregating the two scores for each putative drug–target interaction (row 8 of Tables 1–4). In the benchmark experiments (a), (c) and (d), the aggregated curve mimics or gives a significant improvement over the other two curves. For ion channels (b), leave-one-out on potential target proteins (dashed line) perform slightly better overall than aggregation (dotted line), but both curves represent extremely strong results.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2735674&req=5

Figure 1: PR curves for predicted drug–target interactions using BLMs on four benchmark datasets: (a) enzyme, (b) ion channel, (c) GPCR and (d) nuclear receptor. The solid line is for leave-one-out on potential drugs (row 2 of Tables 1–4), the dashed line for leave-one-out on potential target proteins (row 5 of Tables 1–4) and the dotted line for aggregating the two scores for each putative drug–target interaction (row 8 of Tables 1–4). In the benchmark experiments (a), (c) and (d), the aggregated curve mimics or gives a significant improvement over the other two curves. For ion channels (b), leave-one-out on potential target proteins (dashed line) perform slightly better overall than aggregation (dotted line), but both curves represent extremely strong results.

Mentions: This corresponds to rows 7–9 and 12 in Tables 1–4 and simulates the prediction of missing drug–target interactions in the known bipartite network. Here, we see the significant improvement in AUC and AUPR scores that can be achieved by aggregating the set of prediction scores for the same drug–target interaction (edge) into a global prediction score, also illustrated in the PR curves of Figure 1. We see that in all cases, aggregating scores across the BLM method or across the two methods gives the best AUC and AUPR scores. It is important to note that the idea to aggregate scores via a function m was not used in Yamanishi et al. (2008), so the aggregated results shown in Tables 1–4 for their method, seen alone, are actually an improvement on the results shown in their original article.Fig. 1.


Supervised prediction of drug-target interactions using bipartite local models.

Bleakley K, Yamanishi Y - Bioinformatics (2009)

PR curves for predicted drug–target interactions using BLMs on four benchmark datasets: (a) enzyme, (b) ion channel, (c) GPCR and (d) nuclear receptor. The solid line is for leave-one-out on potential drugs (row 2 of Tables 1–4), the dashed line for leave-one-out on potential target proteins (row 5 of Tables 1–4) and the dotted line for aggregating the two scores for each putative drug–target interaction (row 8 of Tables 1–4). In the benchmark experiments (a), (c) and (d), the aggregated curve mimics or gives a significant improvement over the other two curves. For ion channels (b), leave-one-out on potential target proteins (dashed line) perform slightly better overall than aggregation (dotted line), but both curves represent extremely strong results.
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2735674&req=5

Figure 1: PR curves for predicted drug–target interactions using BLMs on four benchmark datasets: (a) enzyme, (b) ion channel, (c) GPCR and (d) nuclear receptor. The solid line is for leave-one-out on potential drugs (row 2 of Tables 1–4), the dashed line for leave-one-out on potential target proteins (row 5 of Tables 1–4) and the dotted line for aggregating the two scores for each putative drug–target interaction (row 8 of Tables 1–4). In the benchmark experiments (a), (c) and (d), the aggregated curve mimics or gives a significant improvement over the other two curves. For ion channels (b), leave-one-out on potential target proteins (dashed line) perform slightly better overall than aggregation (dotted line), but both curves represent extremely strong results.
Mentions: This corresponds to rows 7–9 and 12 in Tables 1–4 and simulates the prediction of missing drug–target interactions in the known bipartite network. Here, we see the significant improvement in AUC and AUPR scores that can be achieved by aggregating the set of prediction scores for the same drug–target interaction (edge) into a global prediction score, also illustrated in the PR curves of Figure 1. We see that in all cases, aggregating scores across the BLM method or across the two methods gives the best AUC and AUPR scores. It is important to note that the idea to aggregate scores via a function m was not used in Yamanishi et al. (2008), so the aggregated results shown in Tables 1–4 for their method, seen alone, are actually an improvement on the results shown in their original article.Fig. 1.

Bottom Line: We demonstrate the excellent performance of the proposed method in the prediction of four classes of drug-target interaction networks involving enzymes, ion channels, G protein-coupled receptors (GPCRs) and nuclear receptors in human.This enables us to suggest a number of new potential drug-target interactions.An implementation of the proposed algorithm is available upon request from the authors.

View Article: PubMed Central - PubMed

Affiliation: Mines ParisTech, Centre for Computational Biology, Fontainebleau, France. kevbleakley@gmail.com

ABSTRACT

Motivation: In silico prediction of drug-target interactions from heterogeneous biological data is critical in the search for drugs for known diseases. This problem is currently being attacked from many different points of view, a strong indication of its current importance. Precisely, being able to predict new drug-target interactions with both high precision and accuracy is the holy grail, a fundamental requirement for in silico methods to be useful in a biological setting. This, however, remains extremely challenging due to, amongst other things, the rarity of known drug-target interactions.

Results: We propose a novel supervised inference method to predict unknown drug-target interactions, represented as a bipartite graph. We use this method, known as bipartite local models to first predict target proteins of a given drug, then to predict drugs targeting a given protein. This gives two independent predictions for each putative drug-target interaction, which we show can be combined to give a definitive prediction for each interaction. We demonstrate the excellent performance of the proposed method in the prediction of four classes of drug-target interaction networks involving enzymes, ion channels, G protein-coupled receptors (GPCRs) and nuclear receptors in human. This enables us to suggest a number of new potential drug-target interactions.

Availability: An implementation of the proposed algorithm is available upon request from the authors. Datasets and all prediction results are available at http://cbio.ensmp.fr/~yyamanishi/bipartitelocal/.

Show MeSH
Related in: MedlinePlus