Integrating heterogeneous knowledge graphs into drug-drug interaction extraction from the literature.

MOTIVATION: Most of the conventional deep neural network-based methods for drug-drug interaction (DDI) extraction consider only context information around drug mentions in the text. However, human experts use heterogeneous background knowledge about drugs to comprehend pharmaceutical papers and extract relationships between drugs. Therefore, we propose a novel method that simultaneously considers various heterogeneous information for DDI extraction from the literature. RESULTS: We first construct drug representations by conducting the link prediction task on a heterogeneous pharmaceutical knowledge graph (KG) dataset. We then effectively combine the text information of input sentences in the corpus and the information on drugs in the heterogeneous KG (HKG) dataset. Finally, we evaluate our DDI extraction method on the DDIExtraction-2013 shared task dataset. In the experiment, integrating heterogeneous drug information significantly improves the DDI extraction performance, and we achieved an F-score of 85.40%, which results in state-of-the-art performance. We evaluated our method on the DrugProt dataset and improved the performance significantly, achieving an F-score of 77.9%. Further analysis showed that each type of node in the HKG contributes to the performance improvement of DDI extraction, indicating the importance of considering multiple pieces of information. AVAILABILITY AND IMPLEMENTATION: Our code is available at https://github.com/tticoin/HKG-DDIE.git.

Representing a Heterogeneous Pharmaceutical Knowledge-Graph with Textual Information.

We deal with a heterogeneous pharmaceutical knowledge-graph containing textual information built from several databases. The knowledge graph is a heterogeneous graph that includes a wide variety of concepts and attributes, some of which are provided in the form of textual pieces of information which have not been targeted in the conventional graph completion tasks. To investigate the utility of textual information for knowledge graph completion, we generate embeddings from textual descriptions given to heterogeneous items, such as drugs and proteins, while learning knowledge graph embeddings. We evaluate the obtained graph embeddings on the link prediction task for knowledge graph completion, which can be used for drug discovery and repurposing. We also compare the results with existing methods and discuss the utility of the textual information.

Using drug descriptions and molecular structures for drug-drug interaction extraction from literature.

MOTIVATION: Neural methods to extract drug-drug interactions (DDIs) from literature require a large number of annotations. In this study, we propose a novel method to effectively utilize external drug database information as well as information from large-scale plain text for DDI extraction. Specifically, we focus on drug description and molecular structure information as the drug database information. RESULTS: We evaluated our approach on the DDIExtraction 2013 shared task dataset. We obtained the following results. First, large-scale raw text information can greatly improve the performance of extracting DDIs when combined with the existing model and it shows the state-of-the-art performance. Second, each of drug description and molecular structure information is helpful to further improve the DDI performance for some specific DDI types. Finally, the simultaneous use of the drug description and molecular structure information can significantly improve the performance on all the DDI types. We showed that the plain text, the drug description information and molecular structure information are complementary and their effective combination is essential for the improvement. AVAILABILITY AND IMPLEMENTATION: Our code is available at https://github.com/tticoin/DESC_MOL-DDIE.

Discovery of a Slow Tight Binding LPA1 Antagonist (ONO-0300302) for the Treatment of Benign Prostatic Hyperplasia.

Scaffold hopping from the amide group of lead compound ONO-7300243 (1) to a secondary alcohol successfully gave a novel chemotype lysophosphatidic acid receptor 1 (LPA1) antagonist 4. Wash-out experiments using rat isolated urethra showed that compound 4 possesses a tight binding feature to the LPA1 receptor. Further modification of two phenyl groups of 1 to pyrrole and an indane moiety afforded an optimized compound ONO-0300302 (19). Despite its high i.v. clearance, 19 inhibited significantly an LPA-induced increase of intraurethral pressure (IUP) in rat (3 mg/kg, p.o.) and dog (1 mg/kg, p.o.) over 12 h. Binding experiments with [3H]-ONO-0300302 suggest that the observed long duration action is because of the slow tight binding character of 19.

Discovery of an 8-aza-5-thiaProstaglandin E1 analog as a highly selective EP4 receptor agonist.

For the purpose of discovering an orally available EP4 subtype-selective agonist, a series of 8-aza prostaglandin E(1) (PGE(1)) analogs were synthesized and evaluated for their affinity for PGE(2) receptor subtypes. Additionally, the structure-activity relationships of these compounds were studied. Among the tested compounds, the 8-aza PGE(1) analog 6 and 8-aza-5-thiaPGE(1) analog 12 had highly potent EP4 receptor affinity, good functional activity, and excellent subtype-selectivity. Furthermore, these analogs demonstrated good stability in human liver microsomes. As a result, we concluded that these two series of 8-aza PGE(1) analogs could be promising chemical leads for an orally available EP4 subtype-selective agonist.

3-(2-Aminocarbonylphenyl)propanoic acid analogs as potent and selective EP3 receptor antagonists. Part 3: Synthesis, metabolic stability, and biological evaluation of optically active analogs.

A series of 3-(2-aminocarbonylphenyl)propanoic acid analogs possessing the (1R)-1-(3,5-dimethylphenyl)-3-methylbutylamine moiety on the carboxyamide side chain were synthesized and evaluated for their binding affinity for the EP1-4 receptors and their antagonist activity for the EP3 receptor. Rational drug design based on the structure of the metabolites in human liver microsomes led us to the discovery of another series of analogs. Several compounds were further evaluated for their in vivo efficacy in rats after oral administration and also for their pharmacokinetic profiles including in vitro stability in the liver microsomes.

Discovery of novel N-acylsulfonamide analogs as potent and selective EP3 receptor antagonists.

A series of novel N-acylsulfonamide analogs were synthesized and evaluated for their binding affinity and antagonist activity for the EP3 receptor subtype. Representative compounds were also evaluated for their inhibitory effect on PGE(2)-induced uterine contraction in pregnant rats. Among those tested, a series of N-acylbenzenesulfonamide analogs were found to be more potent than the corresponding carboxylic acid analogs in both the in vitro and in vivo evaluations. The structure activity relationships (SAR) are also discussed.

3-(2-Aminocarbonylphenyl)propanoic acid analogs as potent and selective EP3 receptor antagonists. part 2: optimization of the side chains to improve in vitro and in vivo potencies.

A series of 3-[2-{[(3-methyl-1-phenylbutyl)amino]carbonyl}-4-(phenoxymethyl)phenyl]propanoic acid analogs were synthesized and evaluated for their in vitro potency. In most cases, introduction of one or two substituents into the two phenyl moieties resulted in the tendency of an increase or retention of in vitro activities. Several compounds, which showed excellent subtype selectivity, were evaluated for their inhibitory effect against PGE(2)-induced uterine contraction in pregnant rats, which is thought to be mediated by the EP3 receptor subtype. The structure-activity relationships (SARs) are also discussed.

3-(2-Aminocarbonylphenyl)propanoic acid analogs as potent and selective EP3 receptor antagonists. Part 1: discovery and exploration of the carboxyamide side chain.

A series of 3-(2-aminocarbonyl-4-phenoxymethylphenyl)propanoic acid analogs were synthesized and evaluated for their EP3 antagonist activity in the presence of additive serum albumin. Several compounds were biologically evaluated for their in vivo efficacy with respect to the PGE(2)-induced uterine contraction in pregnant rats as well as their pharmacokinetics. The discovery process of these potent and selective EP3 antagonists and their structure activity relationship are also presented.

Discovery of a series of acrylic acids and their derivatives as chemical leads for selective EP3 receptor antagonists.

A series of acrylic acids and their structurally related compounds were evaluated for their binding affinity to four EP receptor subtypes (EP1-4). Starting from the initial hit 3, which was discovered in our in-house library, compounds 4 and 5 were identified as new chemical leads as candidates for further optimization towards a selective EP3 receptor antagonist. The identification process of these compounds and their pharmacokinetic profiles are presented.

Design and synthesis of a selective EP4-receptor agonist. Part 3: 16-phenyl-5-thiaPGE(1) and 9-beta-halo derivatives with improved stability.

To identify a new selective EP4-agonist with improved chemical stability, further chemical modification of those reported previously was continued. We focused our attention on chemical modification of the alpha chain of 3,7-dithiaPGE(1) and selected 5-thiaPGE(1) as a new chemical lead. Introduction of an optimized omega chain to the 5-thiaPG skeleton afforded m-methoxymethyl derivative 33a, which showed the most potent EP4-receptor agonist activity and good subtype-selectivity both in vitro and in vivo. 9beta-HaloPGF derivatives were also synthesized and biologically evaluated in an attempt to block self-degradation of the beta-hydroxyketone moiety. Among these series, and 39b showed potent agonist activity and good subtype-selectivity. Structure-activity relationships (SARs) are also discussed.

Design and synthesis of a selective EP4-receptor agonist. Part 1: discovery of 3,7-dithiaPGE1 derivatives and identification of their omega chains.

Improvement of EP4-receptor selectivity and the agonist activity by introduction of heteroatoms into the alpha chain of PGE1 was investigated. Among the compounds tested, 3,7-dithiaPGE1 4a exhibited good EP4-receptor selectivity and agonist activity. Further modification of the omega chain of 3,7-dithiaPGE1 was performed to improve EP4-receptor selectivity and agonist activity. Of the compounds produced, 16-phenyl-omega-tetranor-3,7-dithiaPGE1 4p possessing moderate EP4-receptor selectivity and agonist activity, was identified as a new chemical lead for further optimization by modification of the aromatic moiety.

Design and synthesis of a selective EP4-receptor agonist. Part 2: 3,7-dithiaPGE1 derivatives with high selectivity.

To identify new highly selective EP4-agonists, further modification of the 16-phenyl moiety of 1 was continued. 16-(3-methoxymethyl)phenyl derivatives 13-(6q) and 16-(3-ethoxymethyl)phenyl derivatives 13-(7e) showed more selectivity and potent agonist activity than 1. 16-(3-methyl-4-hydroxy)phenyl derivative 18-(14e) demonstrated excellent subtype selectivity, while both its receptor affinity and agonist activity were less potent than those of 13-(6q). Structure-activity relationships (SARs) are also discussed.