Meta-QSAR: a large-scale application of meta-learning to drug design and discovery.

We investigate the learning of quantitative structure activity relationships (QSARs) as a case-study of meta-learning. This application area is of the highest societal importance, as it is a key step in the development of new medicines. The standard QSAR learning problem is: given a target (usually a protein) and a set of chemical compounds (small molecules) with associated bioactivities (e.g. inhibition of the target), learn a predictive mapping from molecular representation to activity. Although almost every type of machine learning method has been applied to QSAR learning there is no agreed single best way of learning QSARs, and therefore the problem area is well-suited to meta-learning. We first carried out the most comprehensive ever comparison of machine learning methods for QSAR learning: 18 regression methods, 3 molecular representations, applied to more than 2700 QSAR problems. (These results have been made publicly available on OpenML and represent a valuable resource for testing novel meta-learning methods.) We then investigated the utility of algorithm selection for QSAR problems. We found that this meta-learning approach outperformed the best individual QSAR learning method (random forests using a molecular fingerprint representation) by up to 13%, on average. We conclude that meta-learning outperforms base-learning methods for QSAR learning, and as this investigation is one of the most extensive ever comparisons of base and meta-learning methods ever made, it provides evidence for the general effectiveness of meta-learning over base-learning.

Assessing Risks of Shallow Riparian Groundwater Quality Near an Oil Sands Tailings Pond.

The potential discharge of groundwater contaminated by oil sands process-affected water (OSPW) is a concern for aquatic ecosystems near tailings ponds. Groundwater in the area, but unaffected by OSPW, may contain similar compounds, complicating the assessment of potential ecological impacts. In this study, 177 shallow groundwater samples were collected from riparian areas along the Athabasca River and tributaries proximate to oil sands developments. For "pond-site" samples (71; adjacent to study tailings pond), Canadian aquatic life guidelines were exceeded for 11 of 20 assessed compounds. However, "non-pond" samples (54; not near any tailings pond) provided similar exceedances. Statistical analyses indicate that pond-site and non-pond samples were indistinguishable for all but seven parameters assessed, including salts, many trace metals, and fluorescence profiles of aromatic naphthenic acids (ANA). This suggests that, regarding the tested parameters, groundwater adjacent to the study tailings pond generally poses no greater ecological risk than other nearby groundwaters at this time. Multivariate analyses applied to the groundwater data set separated into 11 smaller zones support this conclusion, but show some variation between zones. Geological and potential OSPW influences could not be distinguished based on major ions and metals concentrations. However, similarities in indicator parameters, namely ANA, F, Mo, Se, and Na-Cl ratio, were noted between a small subset of samples from two pond-site zones and two OSPW samples and two shallow groundwater samples documented as likely OSPW affected. This indicator-based screening suggests that OSPW-affected groundwater may be reaching Athabasca River sediments at a few locations.

Residues of the herbicide glyphosate in riparian groundwater in urban catchments.

The herbicide glyphosate and its putative metabolite aminomethylphosphonic acid (AMPA) have been found in urban streams, but limited information is available on their presence in urban riparian groundwater. Information is also lacking regarding the source of AMPA in these urban settings (glyphosate metabolite or wastewater), and whether, if present, glyphosate residues in urban riparian groundwater contribute significantly to urban streams. Glyphosate and AMPA were detected in shallow riparian groundwater at 4 of 5 stream sites in urban catchments in Canada and each were found in approximately 1 in 10 of the samples overall. Frequency of observations of glyphosate and AMPA varied substantially between sites, from no observations in a National Park near the Town of Jasper Alberta, to observations of both glyphosate and AMPA in more than half of the samples along two short reaches of streams in Burlington, Ontario. In these two catchments, AMPA was correlated with glyphosate, rather than the artificial sweetener acesulfame, suggesting that the AMPA is derived mainly from glyphosate degradation rather than from wastewater sources. Land use, localized dosage history, depth below ground and other factors likely control the occurrence of detectable glyphosate residues in groundwater.

An artificial sweetener and pharmaceutical compounds as co-tracers of urban wastewater in groundwater.

Groundwater in urban areas can be affected by numerous wastewater sources. Distinguishing these sources can facilitate better management of urban water resources and wastewater, and protection of urban aquatic environments. A single wastewater tracer, even if ideal (i.e. low background levels, non-reactive, low detection limits, etc.), would be unable to accomplish this task. Here, we investigated the potential advantages of using a suite of anthropogenic chemicals as co-tracers to distinguish wastewater sources that contribute to groundwater contamination at two urban sites. We considered both relatively ubiquitous and non-ubiquitous tracers in wastewater. At the Jasper (Alberta, Canada) site, concentrations of an artificial sweetener, two pharmaceutical compounds, and a degradate of nicotine in groundwater were strongly correlated as co-tracers. This evidence, along with the similar spatial distributions of these co-tracers could be used to delineate and distinguish a single municipal wastewater plume. At the Barrie (Ontario, Canada) site, there was moderate to strong correlation of the wastewater co-tracers, but local differences in their distributions and in the ratios of their concentrations could be used to infer that mixtures of two or more domestic septic plumes were present in the groundwater at this site. This study demonstrates the benefit of applying a suite of tracers to urban groundwater affected by wastewater contamination. This approach should be applicable at other urban sites.

Chemical fingerprinting of naphthenic acids and oil sands process waters-A review of analytical methods for environmental samples.

This article provides a review of the routine methods currently utilized for total naphthenic acid analyses. There is a growing need to develop chemical methods that can selectively distinguish compounds found within industrially derived oil sands process affected waters (OSPW) from those derived from the natural weathering of oil sands deposits. Attention is thus given to the characterization of other OSPW components such as oil sands polar organic compounds, PAHs, and heavy metals along with characterization of chemical additives such as polyacrylamide polymers and trace levels of boron species. Environmental samples discussed cover the following matrices: OSPW containments, on-lease interceptor well systems, on- and off-lease groundwater, and river and lake surface waters. There are diverse ranges of methods available for analyses of total naphthenic acids. However, there is a need for inter-laboratory studies to compare their accuracy and precision for routine analyses. Recent advances in high- and medium-resolution mass spectrometry, concomitant with comprehensive mass spectrometry techniques following multi-dimensional chromatography or ion-mobility separations, have allowed for the speciation of monocarboxylic naphthenic acids along with a wide range of other species including humics. The distributions of oil sands polar organic compounds, particularly the sulphur containing species (i.e., OxS and OxS2) may allow for distinguishing sources of OSPW. The ratios of oxygen- (i.e., Ox) and nitrogen-containing species (i.e., NOx, and N2Ox) are useful for differentiating organic components derived from OSPW from natural components found within receiving waters. Synchronous fluorescence spectroscopy also provides a powerful screening technique capable of quickly detecting the presence of aromatic organic acids contained within oil sands naphthenic acid mixtures. Synchronous fluorescence spectroscopy provides diagnostic profiles for OSPW and potentially impacted groundwater that can be compared against reference groundwater and surface water samples. Novel applications of X-ray absorption near edge spectroscopy (XANES) are emerging for speciation of sulphur-containing species (both organic and inorganic components) as well as industrially derived boron-containing species. There is strong potential for an environmental forensics application of XANES for chemical fingerprinting of weathered sulphur-containing species and industrial additives in OSPW.

What can we learn from the evolution of protein-ligand interactions to aid the design of new therapeutics?

Efforts to increase affinity in the design of new therapeutic molecules have tended to lead to greater lipophilicity, a factor that is generally agreed to be contributing to the low success rate of new drug candidates. Our aim is to provide a structural perspective to the study of lipophilic efficiency and to compare molecular interactions created over evolutionary time with those designed by humans. We show that natural complexes typically engage in more polar contacts than synthetic molecules bound to proteins. The synthetic molecules also have a higher proportion of unmatched heteroatoms at the interface than the natural sets. These observations suggest that there are lessons to be learnt from Nature, which could help us to improve the characteristics of man-made molecules. In particular, it is possible to increase the density of polar contacts without increasing lipophilicity and this is best achieved early in discovery while molecules remain relatively small.

Automated design of ligands to polypharmacological profiles.

The clinical efficacy and safety of a drug is determined by its activity profile across many proteins in the proteome. However, designing drugs with a specific multi-target profile is both complex and difficult. Therefore methods to design drugs rationally a priori against profiles of several proteins would have immense value in drug discovery. Here we describe a new approach for the automated design of ligands against profiles of multiple drug targets. The method is demonstrated by the evolution of an approved acetylcholinesterase inhibitor drug into brain-penetrable ligands with either specific polypharmacology or exquisite selectivity profiles for G-protein-coupled receptors. Overall, 800 ligand-target predictions of prospectively designed ligands were tested experimentally, of which 75% were confirmed to be correct. We also demonstrate target engagement in vivo. The approach can be a useful source of drug leads when multi-target profiles are required to achieve either selectivity over other drug targets or a desired polypharmacology.

Differential changes in gene expression in rainbow trout hepatocytes exposed to extracts of oil sands process-affected water and the Athabasca River.

The oil sands region of northern Alberta represents the world's largest reserves of bitumen, and the accelerated pace of industrial extraction activity has raised concern about the possible impacts on the Athabasca River and its tributaries. An ecotoxicogenomic study was undertaken on Oncorhynchus mykiss trout hepatocytes exposed to extracts of water samples near the oil sand development area, as well as to oil sands process-affected water (OSPW) extracts using the quantitative reverse transcriptase polymerase chain reaction technique. The expression of the following genes (mRNA) was monitored to track changes in xenobiotic biotransformation (CYP1A1, CYP3A4, glutathione S-transferase, multi-drug resistance transporter), estrogenicity (estrogen receptor and vitellogenin), oxidative stress (superoxide dismutase and metallothionein) and DNA repair activity (DNA ligase). The extent of DNA-aromatic hydrocarbon adducts was also determined in cells by immuno-staining. A comparative analysis of gene expression between the river/lake and OSPW samples revealed that CYP3A4, metallothioneins, DNA ligase and GST genes, were specifically expressed by OSPW. Cells exposed to OSPW, commercial naphthenic acids, and benzo(a)pyrene showed increased polyaromatic hydrocarbon DNA-adducts, as determined by cell immunofluorescence analysis. Other genes were induced by all types of water samples, although the induction potential was stronger in OSPW most of the time (e.g., VTG gene was expressed nearly 15-fold by surface waters from the lake and river samples but increased to a maximum of 31-fold in OSPW). A multivariate discriminant function analysis revealed that the lake and river water samples were well discriminated from the OSPW. The CYP3A4 gene was the most highly expressed gene in cells exposed to OSPW and responded less to the lake or river water in the Athabasca River area. This study identified a suite of gene targets that responded specifically to OSPW extracts, which could serve as toxicogenomic fingerprints of OSPW contamination.

Quantifying the chemical beauty of drugs.

Drug-likeness is a key consideration when selecting compounds during the early stages of drug discovery. However, evaluation of drug-likeness in absolute terms does not reflect adequately the whole spectrum of compound quality. More worryingly, widely used rules may inadvertently foster undesirable molecular property inflation as they permit the encroachment of rule-compliant compounds towards their boundaries. We propose a measure of drug-likeness based on the concept of desirability called the quantitative estimate of drug-likeness (QED). The empirical rationale of QED reflects the underlying distribution of molecular properties. QED is intuitive, transparent, straightforward to implement in many practical settings and allows compounds to be ranked by their relative merit. We extended the utility of QED by applying it to the problem of molecular target druggability assessment by prioritizing a large set of published bioactive compounds. The measure may also capture the abstract notion of aesthetics in medicinal chemistry.

Preliminary fingerprinting of Athabasca oil sands polar organics in environmental samples using electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry.

There is a growing need to develop analytical methods that can distinguish compounds found within industrially derived oil sands process water (OSPW) from those derived from natural weathering of oil sands deposits. This is a difficult challenge as possible leakage beyond tailings pond containments will probably be in the form of mixtures of water-soluble organics that may be similar to those leaching naturally into aquatic environments. We have evaluated the potential of negative ion electrospray ionization high-resolution Fourier transform ion cyclotron resonance mass spectrometry (FTICRMS) for comparing oil sands polar organics from tailing ponds, interceptor wells, groundwater, river and lake surface waters. Principal component analysis was performed for all species observed. which included the O(2) class (often assumed to be monocarbxoylic naphthenic acids) along with a wide range of other species including humic substances in the river and lake samples: O(n) where n=1-16; NO(n) and N(2)O(n) where n=1-13; and O(n)S and O(n)S(2) where n=1-10 and 1-8, respectively. A broad range of species was investigated because classical naphthenic acids can be a small fraction of the 'organics' detected in the polar fraction of OSPW, river water and groundwater. Aquatic toxicity and environmental chemistry are attributed to the total organics (not only the classical naphthenic acids). The distributions of the oil sands polar organics, particularly the sulfur-containing species, O(n)S and O(n)S(2), may have potential for distinguishing sources of OSPW. The ratios of species containing O(n) along with nitrogen-containing species: NO(n), and N(2)O(n), were useful for differentiating organic components derived from OSPW from those found in river and lake waters. Further application of the FTICRMS technique for a diverse range of OSPW of varying ages and composition, as well as the surrounding groundwater wells, may be critical in assessing whether leakage from industrial sources to natural waters is occurring.

Rapid analysis of pharmacology for infectious diseases.

Pandemic, epidemic and endemic infectious diseases are united by a common problem: how do we rapidly and cost-effectively identify potential pharmacological interventions to treat infections? Given the large number of emerging and neglected infectious diseases and the fact that they disproportionately afflict the poorest members of the global society, new ways of thinking are required to developed high productivity discovery systems that can be applied to a larger number of pathogens. The growing availability of parasite genome data provides the basis for developing methods to prioritize, a priori, the potential drug target and pharmacological landscape of an infectious disease. Thus the overall objective of infectious disease informatics is to enable the rapid generation of plausible, novel medical hypotheses of testable pharmacological experiments, by uncovering undiscovered relationships in the wealth of biomedical literature and databases that were collected for other purposes. In particular our goal is to identify potential drug targets present in a pathogen genome and prioritize which pharmacological experiments are most likely to discover drug-like lead compounds rapidly against a pathogen (i.e. which specific compounds and drug targets should be screened, in which assays and where they can be sourced). An integral part of the challenge is the development and integration of methods to predict druggability, essentiality, synthetic lethality and polypharmacology in pathogen genomes, while simultaneously integrating the inevitable issues of chemical tractability and the potential for acquired drug resistance from the start.

An ontology for description of drug discovery investigations.

The paper presents an ontology for the description of Drug Discovery Investigation (DDI).This has been developed through the use of a Robot Scientist "Eve", and in consultation with industry. DDI aims to define the principle entities and the relations in the research and development phase of the drug discovery pipeline. DDI is highly transferable and extendable due to its adherence to accepted standards, and compliance with existing ontology resources. This enables DDI to be integrated with such related ontologies as the Vaccine Ontology, the Advancing Clinico-Genomic Trials on Cancer Master Ontology, etc. DDI is available at http://purl.org/ddi/wikipedia or http://purl.org/ddi/home.

Atomic interactions and profile of small molecules disrupting protein-protein interfaces: the TIMBAL database.

Growing evidence of the possibility of modulating protein-protein interactions with small molecules is opening the door to new approaches and concepts in drug discovery. In this paper, we describe the creation of TIMBAL, a hand-curated database holding an up to date collection of small molecules inhibiting multi-protein complexes. This database has been analysed and profiled in terms of molecular properties. Protein-protein modulators tend to be large lipophilic molecules with few hydrogen bond features. An analysis of TIMBAL's intersection with other structural databases, including CREDO (protein-small molecule from the PDB) and PICCOLO (protein-protein from the PDB) reveals that TIMBAL molecules tend to form mainly hydrophobic interactions with only a few hydrogen bonding contacts. With respect to potency, TIMBAL molecules are slightly less efficient than an average medicinal chemistry hit or lead. The database provides a resource that will allow further insights into the types of molecules favoured by protein interfaces and provide a background to continuing work in this area. Access at http://www-cryst.bioc.cam.ac.uk/timbal.

Structural and functional restraints in the evolution of protein families and superfamilies.

Divergent evolution of proteins reflects both selectively advantageous and neutral amino acid substitutions. In the present article, we examine restraints on sequence, which arise from selectively advantageous roles for structure and function and which lead to the conservation of local sequences and structures in families and superfamilies. We analyse structurally aligned members of protein families and superfamilies in order to investigate the importance of the local structural environment of amino acid residues in the acceptance of amino acid substitutions during protein evolution. We show that solvent accessibility is the most important determinant, followed by the existence of hydrogen bonds from the side-chain to main-chain functions and the nature of the element of secondary structure to which the amino acid contributes. Polar side chains whose hydrogen-bonding potential is satisfied tend to be more conserved than their unsatisfied or non-hydrogen-bonded counterparts, and buried and satisfied polar residues tend to be significantly more conserved than buried hydrophobic residues. Finally, we discuss the importance of functional restraints in the form of interactions of proteins with other macromolecules in assemblies or with substrates, ligands or allosteric regulators. We show that residues involved in such functional interactions are significantly more conserved and have differing amino acid substitution patterns.

Structural bioinformatics mutation analysis reveals genotype-phenotype correlations in von Hippel-Lindau disease and suggests molecular mechanisms of tumorigenesis.

Mutations in the VHL gene lead to von Hippel-Lindau (VHL) disease, a clinically heterogeneous cancer syndrome. Here, we use software and database tools to understand and predict the phenotypes associated with missense mutations in the VHL gene product, pVHL. The protein product pVHL is known to interact with elongin B, elongin C, and the HIF substrate. By analyzing known and predicted interaction sites and predictions of thermodynamic stability change upon mutation, we generate new hypotheses regarding the molecular etiology of renal cell carcinoma (RCC) and pheochromocytoma (PCC) in VHL disease. We find that the molecular causes of RCC and PCC appear to be decoupled. RCC may arise through two distinct mechanisms: disruption of HIF interactions or binding at the elongin B interface. PCC is triggered by mutations which disrupt interactions at the elongin C binding site. These findings have important implications for VHL disease and for nonfamilial RCC, because most cases of clear cell RCC are linked with VHL inactivation. Additionally, predicting effects of genetic variation will be critical as genetic sequencing accelerates; the analytical strategy presented here may elucidate other systems as further data on genetic variation become available.

A structural bioinformatics approach to the analysis of nonsynonymous single nucleotide polymorphisms (nsSNPs) and their relation to disease.

The prediction of the effects of nonsynonymous single nucleotide polymorphisms (nsSNPs) on function depends critically on exploiting all information available on the three-dimensional structures of proteins. We describe software and databases for the analysis of nsSNPs that allow a user to move from SNP to sequence to structure to function. In both structure prediction and the analysis of the effects of nsSNPs, we exploit information about protein evolution, in particular, that derived from investigations on the relation of sequence to structure gained from the study of amino acid substitutions in divergent evolution. The techniques developed in our laboratory have allowed fast and automated sequence-structure homology recognition to identify templates and to perform comparative modeling; as well as simple, robust, and generally applicable algorithms to assess the likely impact of amino acid substitutions on structure and interactions. We describe our strategy for approaching the relationship between SNPs and disease, and the results of benchmarking our approach -- human proteins of known structure and recognized mutation.

Interpretation of injection-withdrawal tracer experiments conducted between two wells in a large single fracture.

Tracer experiments conducted using a flow field established by injecting water into one borehole and withdrawing water from another are often used to establish connections and investigate dispersion in fractured rock. As a result of uncertainty in the uniqueness of existing models used for interpretation, this method has not been widely used to investigate more general transport processes including matrix diffusion or advective solute exchange between mobile and immobile zones of fluid. To explore the utility of the injection-withdrawal method as a general investigative tool and with the intent to resolve the transport processes in a discrete fracture, two tracer experiments were conducted using the injection-withdrawal configuration. The experiments were conducted in a fracture which has a large aperture (>500 microm) and horizontally pervades a dolostone formation. One experiment was conducted in the direction of the hydraulic gradient and the other in the direction opposite to the natural gradient. Two tracers having significantly different values of the free-water diffusion coefficient were used. To interpret the experiments, a hybrid numerical-analytical model was developed which accounts for the arcuate shape of the flow field, advection-dispersion in the fracture, diffusion into the matrix adjacent to the fracture, and the presence of natural flow in the fracture. The model was verified by comparison to a fully analytical solution and to a well-known finite-element model. Interpretation of the tracer experiments showed that when only one tracer, advection-dispersion, and matrix diffusion are considered, non-unique results were obtained. However, by using multiple tracers and by accounting for the presence of natural flow in the fracture, unique interpretations were obtained in which a single value of matrix porosity was estimated from the results of both experiments. The estimate of porosity agrees well with independent measurements of porosity obtained from core samples. This suggests that: (i) the injection-withdrawal method is a viable tool for the investigation of general transport processes provided all relevant experimental conditions are considered and multiple conservative tracers are used; and (ii) for the conditions of the experiments conducted in this study, the dominant mechanism for exchange of solute between the fracture and surrounding medium is matrix diffusion.