Jonathan D. Vessey

Establishing best practise in the application of expert review of mutagenicity under ICH M7.

The ICH M7 guidelines for the assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals allows for the consideration of in silico predictions in place of in vitro studies. This represents a significant advance in the acceptance of (Q)SAR models and has resulted from positive interactions between modellers, regulatory agencies and industry with a shared purpose of developing effective processes to minimise risk. This paper discusses key scientific principles that should be applied when evaluating in silico predictions with a focus on accuracy and scientific rigour that will support a consistent and practical route to regulatory submission.

Toxicological knowledge discovery by mining emerging patterns from toxicity data

Predicting the risk of toxic and environmental effects of chemical compounds is of great importance to all chemical industries [1]. Expert systems have shown success in predicting toxic risk by applying established knowledge of toxicology encoded as a knowledge base of structural alerts and a reasoning model. A disadvantage of expert systems is that developing new structural alerts requires considerable time and effort from domain experts. In order to expedite this process a software tool has been developed that can automatically mine representations of activating features directly from toxicity datasets and present them in an interpretable form. Our knowledge discovery tool applies emerging pattern (EP) mining [2]: a form of association rule mining [3] that is well known to computer science, but is relatively new to chemistry [4]. The EP mining algorithm accepts any data expressed as a series of binary properties, which is divided into two classes, and extracts patterns of those properties that are frequent within the data and are more frequent in one data class compared to the other. By mining emerging patterns from toxicity datasets, encoded as fingerprints of binary descriptors, the tool generates patterns of features that distinguish toxicants from innocuous compounds. These patterns represent potentially activating features of the toxic compounds that may then be used to define new alerts. The knowledge discovery tool has been tested using a public dataset of 3489 mutagens and 2981 non-mutagens, encoded as fingerprints of approximately 2000 functional groups and ring descriptors. EPs were produced and grouped into a number of hierarchical families. Six of the EPs that represented distinct chemical classes were selected for manual inspection by a toxicology expert. Relevant literature was analysed to find a mechanistic rationale for the mined features, which resulted in four new structural alerts for in vitro mutagenicity.

Assessing confidence in predictions made by knowledge-based systems

A new metric, “veracity”, is proposed for assessing the performance of qualitative, reasoning-based prediction systems that takes into account the ability of these systems to express levels of confidence in their predictions. Veracity is shown to be compatible with concordance and it is hoped that it will provide a useful alternative to concordance and other Cooper statistics for the assessment of reasoning-based systems and for comparing them with other types of prediction system. A few datasets for four end points covered by the program, Derek for Windows, have been used to illustrate calculations of veracity. The levels of confidence expressed by Derek for Windows in these examples are shown to carry meaningful information. The approach provides a way of judging how well open predictions (“nothing to report” in Derek for Windows) can support qualified predictions of inactivity.

Self organising hypothesis networks: a new approach for representing and structuring SAR knowledge

BackgroundCombining different sources of knowledge to build improved structure activity relationship models is not easy owing to the variety of knowledge formats and the absence of a common framework to interoperate between learning techniques. Most of the current approaches address this problem by using consensus models that operate at the prediction level. We explore the possibility to directly combine these sources at the knowledge level, with the aim to harvest potentially increased synergy at an earlier stage. Our goal is to design a general methodology to facilitate knowledge discovery and produce accurate and interpretable models.ResultsTo combine models at the knowledge level, we propose to decouple the learning phase from the knowledge application phase using a pivot representation (lingua franca) based on the concept of hypothesis. A hypothesis is a simple and interpretable knowledge unit. Regardless of its origin, knowledge is broken down into a collection of hypotheses. These hypotheses are subsequently organised into hierarchical network. This unification permits to combine different sources of knowledge into a common formalised framework. The approach allows us to create a synergistic system between different forms of knowledge and new algorithms can be applied to leverage this unified model. This first article focuses on the general principle of the Self Organising Hypothesis Network (SOHN) approach in the context of binary classification problems along with an illustrative application to the prediction of mutagenicity.ConclusionIt is possible to represent knowledge in the unified form of a hypothesis network allowing interpretable predictions with performances comparable to mainstream machine learning techniques. This new approach offers the potential to combine knowledge from different sources into a common framework in which high level reasoning and meta-learning can be applied; these latter perspectives will be explored in future work.

A seven-bond coupling, 7J(PP) or through space coupling in the azine diphosphine Z,Z-PPh2CH2C(But)NN(But)CCH2PPh2: crystal structure of Z,Z-P(O)Ph2CH2(But)NN(But)CCH2P(O)Ph2

Abstract Analysis of the 13 C{ 1 H} NMR spectrum of Z,Z -PPh 2 CH 2 C(Bu t )NNC(Bu t )CH 2 PPh 2 ( 1 ) gives second order patterns for the C ipso , C ortho , C meta , P C H 2 and CN carbons caused by a non-zero seven-bond coupling, 7 J (PP). Simulation of the spectra gives 7 J (PP)=4.8 Hz. The corresponding diphosphine dioxide 2 gives first order 13 C{ 1 H} spectra, i.e. 7 J (PP) ∼ 0 Hz. Possible explanations for this unusually large value of 7 J (PP) are discussed. Crystals of 2 are orthorhombic, space group Pbca , with a =1592.2(3), b =1196.1(2), c =1663.0(3) pm and Z =4; final R factor 0.0381 for 1780 observed reflections.

Crystal structure, and variable temperature proton and carbon-13 NMR spectra of the 9-membered ring complex [Cr(CO)4{E,Z-PPh2CH2C(tBu)NNC(tBu)CH2PPh2}]

Abstract Treatment of [Cr(CO) 4 (nbd)] (nbd  norbornadiene) with Z , Z -PPh 2 CH 2 C( t Bu)NNC( t Bu)CH 2 PPh 2 gave the title compound 2b , the structure of which was determined by a single crystal X-ray diffraction study. The proton NMR spectrum of 2b at 308 K and 400 MHz showed that for each of the methylene groups the protons are equivalent, but the spectrum is temperature dependent and at 188 K the slow exchange limiting spectrum is obtained, with all methylene hydrogens chemically non-equivalent; corresponding changes were observed in the 13 C-{ 1 H} NMR spectrum. The behaviour is due to the fluxionality of the 9-membered ring.

The synthesis of endo-3-diphenylphosphino-(1R)-(+)-camphor (L) and some of its complexes with palladium(II), platinum(II) and rhodium(I); crystal structures of L and cis-[PdCl2L2]

Treatment of (1 R )-(+)-camphor with LiBu n , followed by Ph 2 PCl gives, as the main product, exo -3-diphenylphosphino- (1 R )-(+)-camphor ( 3a ) together with some of the corresponding enolate anion 1 . However, on storage the exo - phosphine 3a isomerises to the corresponding endo -phosphine 3b which becomes the main product and was isolated in 70% yield. The crystal structure of 3b was determined and detailed 13 C and proton NMR data are given. In chloroform solution, in the presence of acetic acid as catalyst, the endo -phosphine 3b is partially converted back into the exo -isomer 3a over 2 days. The endo -phosphine 3b (L) with H 2 O 2 gives the corresponding phosphine oxide 3c and with monoclinic sulfur the corresponding phosphine sulfide 3d . With [PdCl 2 (NCPh) 2 ] the endo -phosphine 3b gives [PdCl 2 L 2 ] which exists as a cis-trans mixture 4a and 4b in solution. We have determined the X-ray crystal structure of the cis -form 4b . Treatment of [PtCl 2 (COD)] with L, gives cis -[PtCl 2 L 2 ] ( 4c ) but with [PtCl 2 (NCMe) 2 ] the corresponding trans complex trans -[PtCl 2 L 2 ] ( 4d ) is formed. In compounds 4a-4d the PPh 2 groups are endo . The complexes of type [MCl 2 L 2 ] (MPd or Pt) are also formed by treating the bis-camphorphosphine enolates [M(PPh 2 C 10 H 14 O) 2 ] with HCl. The complexes of type cis -[Pt(CCR) 2 (PPh 2 C 10 H 15 O) 2 ] (RPh or C(Me)CH 2 ) with HCl give exclusively cis -[PtCl 2 L 2 ]. Treatment of [Rh 2 Cl 2 (CO) 4 ] with the endo -phosphine 3b (L) gives trans -[RhCl(CO)L 2 ]. 1 H, 31 P and some 13 C data are given. Crystals of endo -3-diphenylphosphino-(1 R )-(+)-camphor ( 3b ) are orthorhombic, space group P 2 1 2 1 2 1 with a =778.6(1), b =1138.8(1), c =2128.9(3) pm and Z =4, R =0.0329 for 1674 observed reflections. The structure shows that the PPh 2 is endo . Crystals of 4b are orthorhombic, space group P 2 1 2 1 2 1 , with a =1380.3(2), b =1785.1(3), c =1922.8(4) pm and Z =4, R =0.0425 for 4438 observed reflections. The structure shows that the PPh 2 groups are endo and that the phosphines are cis .

Palladium complexes of azines, -diazines and -2-pyridylazines containing (1R)-(+)-camphor or (1R)-()-fenchone groups

Mixed monoazines of types H2CN–NC10H16[C10H16 is a (1R)-(+)-camphor residue (L1), or a (1R)-(–)-fenchone residue (L2)] or Me2CN–NC10H16[C10H16 is a (1R)-(+)-camphor residue (L3)], or α-diazines C10H16N–NCH–CHN–NC10H16[C10H16 is a (1R)-(+)-camphor residue (L4) or a (1R)-(–)-fenchone residue (L5)], or α-2-pyridyl azines C10H16N–NCHC5H4N [C10H16 is a (1R)-(+)-camphor residue (L6) or a (1R)-(–)-fenchone residue (L7)] reacted with Na2PdCl4 to give compounds of type [PdCl2Ln2] for L1, L2 or L3 as monodentate nitrogen-donor ligands and chelated mononuclear complexes of the type [PdCl2Ln] for L4, L5, L6 or L7. The complexes of the monodentate azines L1 and L2 exist in solution as mixtures of isomers differing in the ligating nitrogens. The compounds L1, L2 and L4–L7 also reacted with [Pd2Cl4(PR3)2] to give [PdCl2(PR3)Ln](n= 1, R3= Me2Ph or Me2(C6H4OMe-4); n= 2, R3= Me2Ph, Me2(C6H4OMe-4) or Ph3; n= 4 or 5, R3= Me2Ph; n= 6, R3= Bun3; n= 7, R3= Bun3 or Me2Ph), in which the ligands are monodentate, or [{PdCl2(PR3)}2Ln](Ln= L4–L7, R3= Me2Ph), in which the ligands are bidentate bridging. The phosphine complexes were characterised in solution but were isolated only for [PdCl2(PR3)Ln](Ln= L6, R3= Bun3; L7, R3= Bun3 or Me2Ph; L2, R3= Ph3). Compounds L4, L6 or L7 reacted with the η3-methylallylpalladium complex [{PdCl(η3-CH2CMeCH2)}2] and NH4PF6 to give [Pd(η3-CH2CMeCH2)Ln] PF6(Ln= L4, L6 or L7). The crystal structure of [PdCl2L6]·0.5CH2Cl2 was determined by X-ray diffraction analysis: the crystals are monoclinic, space group P21 with a= 9.5830(13), b= 13.9720(14) and c= 14.726(2)A, β= 97.610(11)° and Z= 4. It shows two molecules of the complex in the asymmetric unit differing only in the torsion angles around the N–N bond.

The stereochemistry of alkyne insertion into Ru–H and Ru–Cl bonds

Proton-coupled 13C NMR studies of vinyl complexes obtained by insertion of MeO2CCCCO2Me into the Ru–H bond of [Ru(CO)2Cl(H)L2](L = PMe2Ph or AsMe2Ph) or one Ru–H bond of [Ru(CO)2H2L2] have shown that the reactions involve trans addition of Ru–H to the alkyne. For the PMe2Ph complexes, selective deuteriation was required to simplify the spectra. In contrast, [Ru(CO)Cl(H)(PMe2Ph)3] reacts by cis addition of Ru–H to the alkyne. Carbonyl substitution in [Ru(CO)2{C(CO2Me)C(CO2Me)H}Cl(PMe2Ph)2] by PMe2Ph leaves the geometry of the vinyl ligand unchanged, so that two (non-interconverting) isomers of [Ru(CO){C(CO2Me)=C(CO2Me)H}Cl(PMe2Ph)3] can be obtained. Whereas trans-[Ru(CO)2Cl2(PMe2Ph)2] combines with MeO2CCCCO2Me by cis addition of Ru–Cl to the alkyne, [Ru(CO)(η2-C2H4)Cl2(PMe2Ph)2] apparently reacts by trans addition, yielding [[graphic omitted]OMe)Cl}Cl(PMe2Ph)2]. Since, however, both reactions appear to involve the same ruthenium intermediate, [Ru(CO)Cl2(PMe2Ph)2], it is possible that the direction of addition is the same in both cases, but that the vinyl ligand can isomerize.

A k‐Nearest Neighbours Approach Using Metabolism‐related Fingerprints to Improve In Silico Metabolite Ranking

The application of biotransformation dictionaries derived by expert evaluation of known metabolic pathways represents one approach to the prediction of both phase I and phase II xenobiotic metabolites. The ranking of metabolites generated by such dictionaries has previously been achieved through the use of qualitative reasoning rules and quantitative probability values. Using the biotransformation dictionary available in the Meteor expert system, we show that metabolite over‐prediction by both of these methods can be reduced by the adoption of a k‐nearest neighbours methodology in which the likelihood of a predicted biotransformation is determined based on comparison of a query chemical with structurally‐similar substrates with known experimental metabolic data which activate the same biotransformation. Optimal performance was achieved when similarity was defined in terms of a combination of two fingerprints, one describing the overall profile of biotransformations a structure can potentially undergo and the other describing the local environment around the predicted site of metabolism for the particular biotransformation under consideration.