Ligand-target prediction by structural network biology using nAnnoLyze

De Novo Prediction of Drug Targets and Candidates by Chemical Similarity-Guided Network-Based Inference

Identifying drug–target interactions is a crucial step in discovering novel drugs and for drug repositioning. Network-based methods have shown great potential thanks to the straightforward integration of information from different sources and the possibility of extracting novel information from the graph topology. However, despite recent advances, there is still an urgent need for efficient and robust prediction methods. Here, we present SimSpread, a novel method that combines network-based inference with chemical similarity. This method employs a tripartite drug–drug–target network constructed from protein–ligand interaction annotations and drug–drug chemical similarity on which a resource-spreading algorithm predicts potential biological targets for both known or failed drugs and novel compounds. We describe small molecules as vectors of similarity indices to other compounds, thereby providing a flexible means to explore diverse molecular representations. We show that our proposed method achieves high prediction performance through multiple cross-validation and time-split validation procedures over a series of datasets. In addition, we demonstrate that our method performed a balanced exploration of both chemical ligand space (scaffold hopping) and biological target space (target hopping). Our results suggest robust and balanced performance, and our method may be useful for predicting drug targets, virtual screening, and drug repositioning.

Computational Methods for Drug Repurposing

The wealth of knowledge and multi-omics data available in drug research has allowed the rise of several computational methods in the drug discovery field, resulting in a novel and exciting strategy called drug repurposing. Drug repurposing consists in finding new applications for existing drugs. Numerous computational methods perform a high-level integration of different knowledge sources to facilitate the discovery of unknown mechanisms. In this chapter, we present a survey of data resources and computational tools available for drug repositioning.

DeepCPI: A Deep Learning-based Framework for Large-scale in silico Drug Screening

Accurate identification of compound–protein interactions (CPIs) in silico may deepen our understanding of the underlying mechanisms of drug action and thus remarkably facilitate drug discovery and development. Conventional similarity- or docking-based computational methods for predicting CPIs rarely exploit latent features from currently available large-scale unlabeled compound and protein data and often limit their usage to relatively small-scale datasets. In the present study, we propose DeepCPI, a novel general and scalable computational framework that combines effective feature embedding (a technique of representation learning) with powerful deep learning methods to accurately predict CPIs at a large scale. DeepCPI automatically learns the implicit yet expressive low-dimensional features of compounds and proteins from a massive amount of unlabeled data. Evaluations of the measured CPIs in large-scale databases, such as ChEMBL and BindingDB, as well as of the known drug–target interactions from DrugBank, demonstrated the superior predictive performance of DeepCPI. Furthermore, several interactions among small-molecule compounds and three G protein-coupled receptor targets (glucagon-like peptide-1 receptor, glucagon receptor, and vasoactive intestinal peptide receptor) predicted using DeepCPI were experimentally validated. The present study suggests that DeepCPI is a useful and powerful tool for drug discovery and repositioning. The source code of DeepCPI can be downloaded from https://github.com/FangpingWan/DeepCPI.

Evaluation of drug efficacy based on the spatial position comparison of drug–target interaction centers

The spatial position and interaction of drugs and their targets is the most important characteristics for understanding a drug’s pharmacological effect, and it could help both in finding new and more precise treatment targets for diseases and in exploring the targeting effects of the new drugs. In this work, we develop a computational pipeline to confirm the spatial interaction relationship of the drugs and their targets and compare the drugs’ efficacies based on the interaction centers. First, we produce a 100-sample set to reconstruct a stable docking model of the confirmed drug–target pairs. Second, we set 5.5 Å as the maximum distance threshold for the drug–amino acid residue atom interaction and construct 3-dimensional interaction surface models. Third, by calculating the spatial position of the 3-dimensional interaction surface center, we develop a comparison strategy for estimating the efficacy of different drug–target pairs. For the 1199 drug–target interactions of the 649 drugs and 355 targets, the drugs that have similar interaction center positions tend to have similar efficacies in disease treatment, especially in the analysis of the 37 targeted relationships between the 15 known anti-cancer drugs and 10 target molecules. Furthermore, the analysis of the unpaired anti-cancer drug and target molecules suggests that there is a potential application for discovering new drug actions using the sampling molecular docking and analyzing method. The comparison of the drug–target interaction center spatial position method better reflect the drug–target interaction situations and could support the discovery of new efficacies among the known anti-cancer drugs.

Predicting protein targets for drug-like compounds using transcriptomics

An expanded chemical space is essential for improved identification of small molecules for emerging therapeutic targets. However, the identification of targets for novel compounds is biased towards the synthesis of known scaffolds that bind familiar protein families, limiting the exploration of chemical space. To change this paradigm, we validated a new pipeline that identifies small molecule-protein interactions and works even for compounds lacking similarity to known drugs. Based on differential mRNA profiles in multiple cell types exposed to drugs and in which gene knockdowns (KD) were conducted, we showed that drugs induce gene regulatory networks that correlate with those produced after silencing protein-coding genes. Next, we applied supervised machine learning to exploit drug-KD signature correlations and enriched our predictions using an orthogonal structure-based screen. As a proof-of-principle for this regimen, top-10/top-100 target prediction accuracies of 26% and 41%, respectively, were achieved on a validation of set 152 FDA-approved drugs and 3104 potential targets. We then predicted targets for 1680 compounds and validated chemical interactors with four targets that have proven difficult to chemically modulate, including non-covalent inhibitors of HRAS and KRAS. Importantly, drug-target interactions manifest as gene expression correlations between drug treatment and both target gene KD and KD of genes that act up- or down-stream of the target, even for relatively weak binders. These correlations provide new insights on the cellular response of disrupting protein interactions and highlight the complex genetic phenotypes of drug treatment. With further refinement, our pipeline may accelerate the identification and development of novel chemical classes by screening compound-target interactions.

Bioactive compounds often disrupt cellular gene expression in ways that are difficult to predict. While the correlation between a cellular response after treatment with a small molecule and the knockdown of its target protein should be simple to establish, in practice this goal has been difficult to achieve. The main challenges are that data are noisy, drugs are not intended to be active in all cell types, and signals from a bona fide target(s) may be obscured by correlations with knockdowns of other proteins in the same pathway(s). Here, we find that a random forest classification model can detect meaningful correlational patterns when gene expression profiles after compound treatment and gene knockdowns in four or more cell lines are compared. When this approach is enriched by a structure-based screen, novel drug-target interactions can be predicted. We then validated new ligand-protein interactions for four difficult targets. Although the initial compounds are not especially potent in vitro, they are capable of disrupting their target pathway in the cell to an extent that generates a significant and characteristic gene expression profile. Collectively, our studies provide insight on cell-level transcriptomic responses to pharmaceutical intervention and the use of these patterns for target identification. In addition, the method provides a novel drug discovery pipeline to test chemistries without a priori knowledge of their target(s).

Ranking Enzyme Structures in the PDB by Bound Ligand Similarity to Biological Substrates

There are numerous applications that use the structures of protein-ligand complexes from the PDB, such as 3D pharmacophore identification, virtual screening, and fragment-based drug design. The structures underlying these applications are potentially much more informative if they contain biologically relevant bound ligands, with high similarity to the cognate ligands. We present a study of ligand-enzyme complexes that compares the similarity of bound and cognate ligands, enabling the best matches to be identified. We calculate the molecular similarity scores using a method called PARITY (proportion of atoms residing in identical topology), which can conveniently be combined to give a similarity score for all cognate reactants or products in the reaction. Thus, we generate a rank-ordered list of related PDB structures, according to the biological similarity of the ligands bound in the structures.

•

We present PARITY, matching atoms in identical topology to gauge ligand similarity

•

Bound-cognate ligand similarity is a useful metric for ranking PDB structures

•

Only 26% of enzyme structures in the PDB have bound-cognate ligand similarity ≥0.7

•

We provide rank-ordered lists of PDBs with the most biologically relevant ligands

Identifying relationships between unrelated pharmaceutical target proteins on the basis of shared active compounds

Aim:

Computational exploration of small-molecule-based relationships between target proteins from different families.

Materials & methods:

Target annotations of drugs and other bioactive compounds were systematically analyzed on the basis of high-confidence activity data.

Results:

A total of 286 novel chemical links were established between distantly related or unrelated target proteins. These relationships involved a total of 1859 bioactive compounds including 147 drugs and 141 targets.

Conclusion:

Computational analysis of large amounts of compounds and activity data has revealed unexpected relationships between diverse target proteins on the basis of compounds they share. These relationships are relevant for drug discovery efforts. Target pairs that we have identified and associated compound information are made freely available.

Relationships between proteins are usually studied by comparing their sequences and functions. However, in addition to biological relationships, chemical links between proteins can also be established by searching for active compounds they share. If proteins have active compounds in common, they are likely to interact with small molecules in similar ways, which provides important clues for drug discovery. Therefore, we have systematically searched for unexpected compound-based relationships between proteins.

Shown here are exemplary small molecules that are active against two targets with different functions. Thus, these compounds establish an unexpected chemical/ligand-binding relationship between these targets.

A Multilayer Network Approach for Guiding Drug Repositioning in Neglected Diseases

Drug development for neglected diseases has been historically hampered due to lack of market incentives. The advent of public domain resources containing chemical information from high throughput screenings is changing the landscape of drug discovery for these diseases. In this work we took advantage of data from extensively studied organisms like human, mouse, E. coli and yeast, among others, to develop a novel integrative network model to prioritize and identify candidate drug targets in neglected pathogen proteomes, and bioactive drug-like molecules. We modeled genomic (proteins) and chemical (bioactive compounds) data as a multilayer weighted network graph that takes advantage of bioactivity data across 221 species, chemical similarities between 1.7 10⁵ compounds and several functional relations among 1.67 10⁵ proteins. These relations comprised orthology, sharing of protein domains, and shared participation in defined biochemical pathways. We showcase the application of this network graph to the problem of prioritization of new candidate targets, based on the information available in the graph for known compound-target associations. We validated this strategy by performing a cross validation procedure for known mouse and Trypanosoma cruzi targets and showed that our approach outperforms classic alignment-based approaches. Moreover, our model provides additional flexibility as two different network definitions could be considered, finding in both cases qualitatively different but sensible candidate targets. We also showcase the application of the network to suggest targets for orphan compounds that are active against Plasmodium falciparum in high-throughput screens. In this case our approach provided a reduced prioritization list of target proteins for the query molecules and showed the ability to propose new testable hypotheses for each compound. Moreover, we found that some predictions highlighted by our network model were supported by independent experimental validations as found post-facto in the literature.

Neglected tropical diseases are human infectious diseases that are often associated with poverty. Historically, lack of interest from the pharmaceutical industry resulted in the lack of good drugs to combat the majority of the pathogens that cause these diseases. Recently, the availability of open chemical information has increased with the advent of public domain chemical resources and the release of data from high throughput screening assays. Our aim in this work was to make use of data from extensively studied organisms like human, mouse, E. coli and yeast, among others, to prioritize and identify candidate drug targets in neglected pathogen proteomes, and drug-like bioactive molecules to foster drug development against neglected diseases. Our approach to the problem relied on applying bioinformatics and computational biology strategies to model large datasets spanning complete proteomes and extensive chemical information from publicly available sources. As a result, we were able to prioritize drug targets and identify potential targets for orphan bioactive drugs.

Predicting Drug-Target Interactions With Multi-Information Fusion

Identifying potential associations between drugs and targets is a critical prerequisite for modern drug discovery and repurposing. However, predicting these associations is difficult because of the limitations of existing computational methods. Most models only consider chemical structures and protein sequences, and other models are oversimplified. Moreover, datasets used for analysis contain only true-positive interactions, and experimentally validated negative samples are unavailable. To overcome these limitations, we developed a semi-supervised based learning framework called NormMulInf through collaborative filtering theory by using labeled and unlabeled interaction information. The proposed method initially determines similarity measures, such as similarities among samples and local correlations among the labels of the samples, by integrating biological information. The similarity information is then integrated into a robust principal component analysis model, which is solved using augmented Lagrange multipliers. Experimental results on four classes of drug-target interaction networks suggest that the proposed approach can accurately classify and predict drug-target interactions. Part of the predicted interactions are reported in public databases. The proposed method can also predict possible targets for new drugs and can be used to determine whether atropine may interact with alpha1B- and beta1- adrenergic receptors. Furthermore, the developed technique identifies potential drugs for new targets and can be used to assess whether olanzapine and propiomazine may target 5HT2B. Finally, the proposed method can potentially address limitations on studies of multitarget drugs and multidrug targets.

Release of 50 new, drug-like compounds and their computational target predictions for open source anti-tubercular drug discovery

As a follow up to the antimycobacterial screening exercise and the release of GSK´s first Tres Cantos Antimycobacterial Set (TCAMS-TB), this paper presents the results of a second antitubercular screening effort of two hundred and fifty thousand compounds recently added to the GSK collection. The compounds were further prioritized based on not only antitubercular potency but also on physicochemical characteristics. The 50 most attractive compounds were then progressed for evaluation in three different predictive computational biology algorithms based on structural similarity or GSK historical biological assay data in order to determine their possible mechanisms of action. This effort has resulted in the identification of novel compounds and their hypothesized targets that will hopefully fuel future TB drug discovery and target validation programs alike.