Library Enhancement through the Wisdom of Crowds

Molecular Similarity Perception Based on Machine-Learning Models

Molecular similarity is an impressively broad topic with many implications in several areas of chemistry. Its roots lie in the paradigm that ‘similar molecules have similar properties’. For this reason, methods for determining molecular similarity find wide application in pharmaceutical companies, e.g., in the context of structure-activity relationships. The similarity evaluation is also used in the field of chemical legislation, specifically in the procedure to judge if a new molecule can obtain the status of orphan drug with the consequent financial benefits. For this procedure, the European Medicines Agency uses experts’ judgments. It is clear that the perception of the similarity depends on the observer, so the development of models to reproduce the human perception is useful. In this paper, we built models using both 2D fingerprints and 3D descriptors, i.e., molecular shape and pharmacophore descriptors. The proposed models were also evaluated by constructing a dataset of pairs of molecules which was submitted to a group of experts for the similarity judgment. The proposed machine-learning models can be useful to reduce or assist human efforts in future evaluations. For this reason, the new molecules dataset and an online tool for molecular similarity estimation have been made freely available.

Individual and collective human intelligence in drug design: evaluating the search strategy

In recent years, individual and collective human intelligence, defined as the knowledge, skills, reasoning and intuition of individuals and groups, have been used in combination with computer algorithms to solve complex scientific problems. Such approach was successfully used in different research fields such as: structural biology, comparative genomics, macromolecular crystallography and RNA design. Herein we describe an attempt to use a similar approach in small-molecule drug discovery, specifically to drive search strategies of de novo drug design. This is assessed with a case study that consists of a series of public experiments in which participants had to explore the huge chemical space in silico to find predefined compounds by designing molecules and analyzing the score associate with them. Such a process may be seen as an instantaneous surrogate of the classical design-make-test cycles carried out by medicinal chemists during the drug discovery hit to lead phase but not hindered by long synthesis and testing times. We present first findings on (1) assessing human intelligence in chemical space exploration, (2) comparing individual and collective human intelligence performance in this task and (3) contrasting some human and artificial intelligence achievements in de novo drug design.

A computational approach yields selective inhibitors of human excitatory amino acid transporter 2 (EAAT2)

Excitatory amino acid transporters (EAATs) represent a protein family that is an emerging drug target with great therapeutic potential for managing central nervous system disorders characterized by dysregulation of glutamatergic neurotransmission. As such, it is of significant interest to discover selective modulators of EAAT2 function. Here, we applied computational methods to identify specific EAAT2 inhibitors. Utilizing a homology model of human EAAT2, we identified a binding pocket at the interface of the transport and trimerization domain. We next conducted a high-throughput virtual screen against this site and identified a selective class of EAAT2 inhibitors that were tested in glutamate uptake and whole-cell electrophysiology assays. These compounds represent potentially useful pharmacological tools suitable for further exploration of the therapeutic potential of EAAT2 and may provide molecular insights into mechanisms of allosteric modulation for glutamate transporters.

An Analysis of Different Components of a High-Throughput Screening Library

Since many projects at pharmaceutical organizations get their start from a high-throughput screening (HTS) campaign, improving the quality of the HTS deck can improve the likelihood of discovering a high-quality lead molecule that can be progressed to a drug candidate. Over the past decade, Janssen has implemented several strategies for external compound acquisition to augment the screening deck beyond the chemical space and number of molecules synthesized for internal projects. In this report, we analyzed the performance of each of those compound collections in the screening campaigns performed internally within Janssen during the last five years. We classified the screening library into two broad categories: Internal and External. The comparison of the performance of these sets of libraries was done by considering the primary, confirmation, and dose response hit rates. Our analysis revealed that Internal compounds (resulting from numerous medicinal chemistry efforts against diverse protein targets) have higher average confirmation hit rates than External ones; however, actives from both categories show similar probabilities of hitting multiple distinct targets. We also investigated the property landscape of both sets of libraries to identify the key elements which make a difference in these categories of compounds. From this analysis, Janssen aims to understand the descriptor landscape of the compounds with the highest hit rates and to use them for improving its future acquisition strategies as well as to inform our plating strategy.

Protocols for the Design of Kinase-focused Compound Libraries

Protocols for the design of kinase-focused compound libraries are presented. Kinase-focused compound libraries can be differentiated based on the design goal. Depending on whether the library should be a discovery library specific for one particular kinase, a general discovery library for multiple distinct kinase projects, or even phenotypic screening, there exists today a variety of in silico methods to design candidate compound libraries. We address the following scenarios: 1) Datamining of SAR databases and kinase focused vendor catalogues; 2) Predictions and virtual screening; 3) Structure-based design of combinatorial kinase inhibitors; 4) Design of covalent kinase inhibitors; 5) Design of macrocyclic kinase inhibitors; and 6) Design of allosteric kinase inhibitors and activators.

Industrial medicinal chemistry insights: neuroscience hit generation at Janssen

The role of medicinal chemistry has changed over the past 10 years. Chemistry had become one step in a process; funneling the output of high-throughput screening (HTS) on to the next stage. The goal to identify the ideal clinical compound remains, but the means to achieve this have changed. Modern medicinal chemistry is responsible for integrating innovation throughout early drug discovery, including new screening paradigms, computational approaches, novel synthetic chemistry, gene-family screening, investigating routes of delivery, and so on. In this Foundation Review, we show how a successful medicinal chemistry team has a broad impact and requires multidisciplinary expertise in these areas.

mQC: A Heuristic Quality-Control Metric for High-Throughput Drug Combination Screening

Quality control (QC) metrics are critical in high throughput screening (HTS) platforms to ensure reliability and confidence in assay data and downstream analyses. Most reported HTS QC metrics are designed for plate level or single well level analysis. With the advent of high throughput combination screening there is a need for QC metrics that quantify the quality of combination response matrices. We introduce a predictive, interpretable, matrix-level QC metric, mQC, based on a mix of data-derived and heuristic features. mQC accurately reproduces the expert assessment of combination response quality and correctly identifies unreliable response matrices that can lead to erroneous or misleading characterization of synergy. When combined with the plate-level QC metric, Z', mQC provides a more appropriate determination of the quality of a drug combination screen. Retrospective analysis on a number of completed combination screens further shows that mQC is able to identify problematic screens whereas plate-level QC was not able to. In conclusion, our data indicates that mQC is a reliable QC filter that can be used to identify problematic drug combinations matrices and prevent further analysis on erroneously active combinations as well as for troubleshooting failed screens. The R source code of mQC is available at http://matrix.ncats.nih.gov/mQC.

Highly efficient and selective phosphorylation of amino acid derivatives and polyols catalysed by 2-aryl-4-(dimethylamino)pyridine-N-oxides--towards kinase-like reactivity

The chemoselective phosphorylation of hydroxyl containing amino acid derivatives and polyols by phosphoryl chlorides catalyzed by 2-aryl-4-(dimethylamino)pyridine-N-oxides is described.

A new modeling approach for quantifying expert opinion in the drug discovery process

Expert opinion plays an important role when choosing clusters of chemical compounds for further investigation. Often, the process by which the clusters are assigned to the experts for evaluation, the so-called selection process, and the qualitative ratings given by the experts to the clusters (chosen/not chosen) need to be jointly modeled to avoid bias. This approach is referred to as the joint modeling approach. However, misspecifying the selection model may impact the estimation and inferences on parameters in the rating model, which are of most scientific interest. We propose to incorporate the selection process into the analysis by adding a new set of random effects to the rating model and, in this way, avoid the need to model it parametrically. This approach is referred to as the combined model approach. Through simulations, the performance of the combined and joint models was compared in terms of bias and confidence interval coverage. The estimates from the combined model were nearly unbiased, and the derived confidence intervals had coverage probability around 95% in all scenarios considered. In contrast, the estimates from the joint model were severely biased under some form of misspecification of the selection model, and fitting the model was often numerically challenging. The results show that the combined model may offer a safer alternative on which to base inferences when there are doubts about the validity of the selection model. Importantly, thanks to its greater numerical stability, the combined model may outperform the joint model even when the latter is correctly specified.

Computational prediction and validation of an expert's evaluation of chemical probes

In a decade with over half a billion dollars of investment, more than 300 chemical probes have been identified to have biological activity through NIH funded screening efforts. We have collected the evaluations of an experienced medicinal chemist on the likely chemistry quality of these probes based on a number of criteria including literature related to the probe and potential chemical reactivity. Over 20% of these probes were found to be undesirable. Analysis of the molecular properties of these compounds scored as desirable suggested higher pKa, molecular weight, heavy atom count, and rotatable bond number. We were particularly interested whether the human evaluation aspect of medicinal chemistry due diligence could be computationally predicted. We used a process of sequential Bayesian model building and iterative testing as we included additional probes. Following external validation of these methods and comparing different machine learning methods, we identified Bayesian models with accuracy comparable to other measures of drug-likeness and filtering rules created to date.

Screening and hit evaluation of a chemical library against blood-stage Plasmodium falciparum

Background

In view of the need to continuously feed the pipeline with new anti-malarial agents adapted to differentiated and more stringent target product profiles (e.g., new modes of action, transmission-blocking activity or long-duration chemo-protection), a chemical library consisting of more than 250,000 compounds has been evaluated in a blood-stage Plasmodium falciparum growth inhibition assay and further assessed for chemical diversity and novelty.

Methods

The selection cascade used for the triaging of hits from the chemical library started with a robust three-step in vitro assay followed by an in silico analysis of the resulting confirmed hits. Upon reaching the predefined requirements for selectivity and potency, the set of hits was subjected to computational analysis to assess chemical properties and diversity. Furthermore, known marketed anti-malarial drugs were co-clustered acting as ‘signposts’ in the chemical space defined by the hits. Then, in cerebro evaluation of the chemical structures was performed to identify scaffolds that currently are or have been the focus of anti-malarial medicinal chemistry programmes. Next, prioritization according to relaxed physicochemical parameters took place, along with the search for structural analogues. Ultimately, synthesis of novel chemotypes with desired properties was performed and the resulting compounds were subsequently retested in a P. falciparum growth inhibition assay.

Results

This screening campaign led to a 1.25% primary hit rate, which decreased to 0.77% upon confirmatory repeat screening. With the predefined potency (EC₅₀ < 1 μM) and selectivity (SI > 10) criteria, 178 compounds progressed to the next steps where chemical diversity, physicochemical properties and novelty assessment were taken into account. This resulted in the selection of 15 distinct chemical series.

Conclusion

A selection cascade was applied to prioritize hits resulting from the screening of a medium-sized chemical library against blood-stage P. falciparum. Emphasis was placed on chemical novelty whereby computational clustering, data mining of known anti-malarial chemotypes and the application of relaxed physicochemical filters, were key to the process. This led to the selection of 15 chemical series from which ten confirmed their activity when newly synthesized sample were tested.

The use of 2D fingerprint methods to support the assessment of structural similarity in orphan drug legislation

BACKGROUND

In the European Union, medicines are authorised for some rare disease only if they are judged to be dissimilar to authorised orphan drugs for that disease. This paper describes the use of 2D fingerprints to show the extent of the relationship between computed levels of structural similarity for pairs of molecules and expert judgments of the similarities of those pairs. The resulting relationship can be used to provide input to the assessment of new active compounds for which orphan drug authorisation is being sought.

RESULTS

143 experts provided judgments of the similarity or dissimilarity of 100 pairs of drug-like molecules from the DrugBank 3.0 database. The similarities of these pairs were also computed using BCI, Daylight, ECFC4, ECFP4, MDL and Unity 2D fingerprints. Logistic regression analyses demonstrated a strong relationship between the human and computed similarity assessments, with the resulting regression models having significant predictive power in experiments using data from submissions of orphan drug medicines to the European Medicines Agency. The BCI fingerprints performed best overall on the DrugBank dataset while the BCI, Daylight, ECFP4 and Unity fingerprints performed comparably on the European Medicines Agency dataset.

CONCLUSIONS

Measures of structural similarity based on 2D fingerprints can provide a useful source of information for the assessment of orphan drug status by regulatory authorities.

The open access malaria box: a drug discovery catalyst for neglected diseases

Historically, one of the key problems in neglected disease drug discovery has been identifying new and interesting chemotypes. Phenotypic screening of the malaria parasite, Plasmodium falciparum has yielded almost 30,000 submicromolar hits in recent years. To make this collection more accessible, a collection of 400 chemotypes has been assembled, termed the Malaria Box. Half of these compounds were selected based on their drug-like properties and the others as molecular probes. These can now be requested as a pharmacological test set by malaria biologists, but importantly by groups working on related parasites, as part of a program to make both data and compounds readily available. In this paper, the analysis and selection methodology and characteristics of the compounds are described.

Social networks to biological networks: systems biology of Mycobacterium tuberculosis

Contextualizing relevant information to construct a network that represents a given biological process presents a fundamental challenge in the network science of biology. The quality of network for the organism of interest is critically dependent on the extent of functional annotation of its genome. Mostly the automated annotation pipelines do not account for unstructured information present in volumes of literature and hence large fraction of genome remains poorly annotated. However, if used, this information could substantially enhance the functional annotation of a genome, aiding the development of a more comprehensive network. Mining unstructured information buried in volumes of literature often requires manual intervention to a great extent and thus becomes a bottleneck for most of the automated pipelines. In this review, we discuss the potential of scientific social networking as a solution for systematic manual mining of data. Focusing on Mycobacterium tuberculosis, as a case study, we discuss our open innovative approach for the functional annotation of its genome. Furthermore, we highlight the strength of such collated structured data in the context of drug target prediction based on systems level analysis of pathogen.

Broad coverage of commercially available lead-like screening space with fewer than 350,000 compounds

In establishing what we propose is the globe's highest quality collection of available screening compounds, it is convincingly shown that the globe's pool of such compounds is extremely shallow and can be represented by fewer than 350,000 compounds. To support our argument, we discuss and fully disclose our extensive battery of functional group filters. We discuss the use of PAINS filters and also show the effect of similarity exclusion on structure-activity relationships. We show why limited analogue representation requires screening at higher concentrations to capture hit classes for difficult targets that otherwise may be prosecuted unsuccessfully. We construct our arguments in a structurally focused manner to be most useful to medicinal chemists, the key players in drug discovery.