Data mining of iron(II) and iron(III) bond-valence parameters, and their relevance for macromolecular crystallography.

The bond-valence model is a reliable way to validate assumed oxidation states based on structural data. It has successfully been employed for analyzing metal-binding sites in macromolecule structures. However, inconsistent results for heme-based structures suggest that some widely used bond-valence R0 parameters may need to be adjusted in certain cases. Given the large number of experimental crystal structures gathered since these initial parameters were determined and the similarity of binding sites in organic compounds and macromolecules, the Cambridge Structural Database (CSD) is a valuable resource for refining metal-organic bond-valence parameters. R0 bond-valence parameters for iron(II), iron(III) and other metals have been optimized based on an automated processing of all CSD crystal structures. Almost all R0 bond-valence parameters were reproduced, except for iron-nitrogen bonds, for which distinct R0 parameters were defined for two observed subpopulations, corresponding to low-spin and high-spin states, of iron in both oxidation states. The significance of this data-driven method for parameter discovery, and how the spin state affects the interpretation of heme-containing proteins and iron-binding sites in macromolecular structures, are discussed.

Hydrogen bonding at C=Se acceptors in selenoureas, selenoamides and selones.

In recent years there has been considerable interest in chalcogen and hydrogen bonding involving Se atoms, but a general understanding of their nature and behaviour has yet to emerge. In the present work, the hydrogen-bonding ability and nature of Se atoms in selenourea derivatives, selenoamides and selones has been explored using analysis of the Cambridge Structural Database and ab initio calculations. In the CSD there are 70 C=Se structures forming hydrogen bonds, all of them selenourea derivatives or selenoamides. Analysis of intramolecular geometries and ab initio partial charges show that this bonding stems from resonance-induced C(δ+)=Se(δ-) dipoles, much like hydrogen bonding to C=S acceptors. C=Se acceptors are in many respects similar to C=S acceptors, with similar vdW-normalized hydrogen-bond lengths and calculated interaction strengths. The similarity between the C=S and C=Se acceptors for hydrogen bonding should inform and guide the use of C=Se in crystal engineering.

The Cambridge Structural Database in retrospect and prospect.

The Cambridge Crystallographic Data Centre (CCDC) was established in 1965 to record numerical, chemical and bibliographic data relating to published organic and metal-organic crystal structures. The Cambridge Structural Database (CSD) now stores data for nearly 700,000 structures and is a comprehensive and fully retrospective historical archive of small-molecule crystallography. Nearly 40,000 new structures are added each year. As X-ray crystallography celebrates its centenary as a subject, and the CCDC approaches its own 50th year, this article traces the origins of the CCDC as a publicly funded organization and its onward development into a self-financing charitable institution. Principally, however, we describe the growth of the CSD and its extensive associated software system, and summarize its impact and value as a basis for research in structural chemistry, materials science and the life sciences, including drug discovery and drug development. Finally, the article considers the CCDC's funding model in relation to open access and open data paradigms.

Hydrogen-bond landscapes, geometry and energetics of squaric acid and its mono- and dianions: a Cambridge Structural Database, IsoStar and computational study.

As part of a programme of work to extend central-group coverage in the Cambridge Crystallographic Data Centre's (CCDC) IsoStar knowledge base of intermolecular interactions, we have studied the hydrogen-bonding abilities of squaric acid (H2SQ) and its mono- and dianions (HSQ(-) and SQ(2-)) using the Cambridge Structural Database (CSD) along with dispersion-corrected density functional theory (DFT-D) calculations for a range of hydrogen-bonded dimers. The -OH and -C=O groups of H2SQ, HSQ(-) and SQ(2-) are potent donors and acceptors, as indicated by their hydrogen-bond geometries in available crystal structures in the CSD, and by the attractive energies calculated for their dimers with acetone and methanol, which were used as model acceptors and donors. The two anions have sufficient examples in the CSD for their addition as new central groups in IsoStar. It is also shown that charge- and resonance-assisted hydrogen bonds involving H2SQ and HSQ(-) are similar in strength to those made by carboxylate COO(-) acceptors, while hydrogen bonds made by the dianion SQ(2-) are somewhat stronger. The study reinforces the value of squaric acid and its anions as cocrystal formers and their actual and potential importance as isosteric replacements for carboxylic acid and carboxylate functions.

Role of chloroform and dichloromethane solvent molecules in crystal packing: an interaction propensity study.

Using the Cambridge Structural Database (CSD), it is shown that the acidic C-H donors of chloroform and dichloromethane, respectively, form hydrogen bonds with N, O, S, halides or carbon-bound halogens in 82% and 77% of structures in which such interactions can occur. This hydrogen-bond potency is retained to a significant degree even in the presence of the more conventional O-H and N-H donors. The hydrogen-bond propensities exhibited by the C-H protons in CHCl3 and CH2Cl2 are similar to those of the acetylenic C-C≡C-H proton. However, involvement of the Cl atoms of CHCl3 and CH2Cl2 in non-bonded interactions is rather limited: the propensities for formation of (O or N)-H...Cl bonds are only 6% in both cases, while the propensities for the formation of halogen-halogen bonds is generally < 15%, with only Cl...Br interactions having slightly higher values. While C(phenyl)-H...Cl interactions are commonly observed, they are of low propensity and have distances at the upper end of the van der Waals limit. We conclude that the acidic C-H protons in chloroform and dichloromethane solvent molecules play a clear role in the involvement of these molecules in molecular aggregation in crystal structures, and this is exemplified by hydrogen-bond predictions made using the statistical propensity tool which is now part of the CSD system.

The versatile role of the ethynyl group in crystal packing: an interaction propensity study.

It is well documented that the ethynyl group can act as a hydrogen-bond donor via its acidic C-H, and as a hydrogen-bond acceptor via the triple-bond π-density. Using the Cambridge Structural Database (CSD), it is shown that C-C≡C-H forms hydrogen bonds to N, O, S or halogens in 74% of structures in which these bonds can form. Additionally, the ethynyl group forms C-H···π interactions with itself or with phenyl groups in 23% of structures and accepts hydrogen bonds from O-H, N-H or C(aromatic)-H in 47% of structures where such bonds are possible. Overall, C-C≡C-H acts as a donor or acceptor in 87% of structures in which it occurs. These propensities for hydrogen-bond formation have been determined using quite tight geometrical constraints, and many more ethynyl groups form interactions with only slight relaxations of these constraints. We conclude that the ethynyl group makes crucial contributions to molecular aggregation in crystal structures, and this is exemplified by hydrogen-bond predictions for specific structures made using the statistical propensity tool now available in CSD system software.

Geometry and conformation of cyclopropane derivatives having σ-acceptor and σ-donor substituents: a theoretical and crystal structure database study.

The structures of cyclopropane rings which carry σ-acceptor or σ-donor substituents have been studied using density-functional theory (DFT), and mean bond lengths and conformational parameters retrieved from the Cambridge Structural Database. It is confirmed that σ-acceptor substituents, e.g. halogens, generate positive ring bond-length asymmetry in which there is lengthening of the distal bond (opposite to the point of substitution), and shorterning of the two vicinal bonds. This is due to withdrawal of electron density from the cyclopropane 1e'' orbitals, which are bonding for the distal bond and antibonding for the vicinal bonds. For σ-donor substituents such as SiH(3) or Si(CH(3))(3), the DFT and crystal structure data show negative ring bond-length asymmetry (distal bond shortened, vicinal bonds lengthened), owing to electron donation into the 4e' ring orbital, which are also bonding for the distal bond and antibonding for the vicinal bonds. The results also show that -OH substituents induce weak positive asymmetry, but that the effects of methyl or amino substituents are either non-existent or extremely small, certainly too small to measure using crystal structure information.

The hydrogen bond environments of 1H-tetrazole and tetrazolate rings: the structural basis for tetrazole-carboxylic acid bioisosterism.

Bioisosterism involving replacement of a carboxylic acid substituent by 1H-tetrazole, yielding deprotonated carboxylate and tetrazolate under physiological conditions, is a well-known synthetic strategy in medicinal chemistry. To improve our overall understanding of bioisosterism, we have used this example to study the geometrical and energetic aspects of the functional group replacement. Specifically, we use crystal structure informatics and high-level ab initio calculations to study the hydrogen bond (H-bond) energy landscapes of the protonated and deprotonated bioisosteric pairs. Each pair exhibits very similar H-bond environments in crystal structures retrieved from the CSD, and the attractive energies of these H-bonds are also very similar. However, by comparison with -COOH and -COO(-), the H-bond environments around 1H-tetrazole and tetrazolate substituents extend further, by about 1.2 Å, from the core of the connected molecule. Analysis of pairs of PDB structures containing ligands which differ only in having a tetrazole or a carboxyl substituent and which are bound to the same protein indicates that the protein binding site must flex sufficiently to form strong H-bonds to either substituent. A survey of DrugBank shows a rather small number of tetrazole-containing drugs in the 'approved' and 'experimental' drug sections of that database.

Conformation and geometry of cyclopropane rings having π-acceptor substituents: a theoretical and database study.

The 3e' orbitals of cyclopropane have the correct symmetry to interact with low-lying unoccupied orbitals of π-acceptor substituents and maximum overlap occurs when the two orbital systems are parallel, i.e. when the π-acceptor bisects the ring in projection down the substituent bond. Since the cyclopropyl group is a common component of active pharmaceutical and agrochemical ingredients, it is important that these strong conjugative interactions are well modelled by computational techniques, and clearly represented in experimental crystal structures. Here we show that torsion angle distributions derived from crystal structure data in the Cambridge Structural Database are in excellent correspondence with torsional energy profiles computed using density functional theory (DFT) for a range of substituents: -COOR, -CONR(2), -NO(2), vinyl and phenyl. We also show that crystal structure information is invaluable in modelling conformations of compounds that contain multiply substituted rings, where steric interactions require some substituents to adopt energetically disfavoured conformations. Further, conjugative interactions with π-acceptors lead to significant asymmetry in the cyclopropane ring bond lengths and again the experimental and computational results are in excellent agreement. Such asymmetry effects are additive, and this explains bond-length variations in cyclopropane rings bearing two or more π-acceptor substituents.

New software for statistical analysis of Cambridge Structural Database data.

A collection of new software tools is presented for the analysis of geometrical, chemical and crystallographic data from the Cambridge Structural Database (CSD). This software supersedes the program Vista. The new functionality is integrated into the program Mercury in order to provide statistical, charting and plotting options alongside three-dimensional structural visualization and analysis. The integration also permits immediate access to other information about specific CSD entries through the Mercury framework, a common requirement in CSD data analyses. In addition, the new software includes a range of more advanced features focused towards structural analysis such as principal components analysis, cone-angle correction in hydrogen-bond analyses and the ability to deal with topological symmetry that may be exhibited in molecular search fragments.

Applications of the Cambridge Structural Database in chemical education.

The Cambridge Structural Database (CSD) is a vast and ever growing compendium of accurate three-dimensional structures that has massive chemical diversity across organic and metal-organic compounds. For these reasons, the CSD is finding significant uses in chemical education, and these applications are reviewed. As part of the teaching initiative of the Cambridge Crystallographic Data Centre (CCDC), a teaching subset of more than 500 CSD structures has been created that illustrate key chemical concepts, and a number of teaching modules have been devised that make use of this subset in a teaching environment. All of this material is freely available from the CCDC website, and the subset can be freely viewed and interrogated using WebCSD, an internet application for searching and displaying CSD information content. In some cases, however, the complete CSD System is required for specific educational applications, and some examples of these more extensive teaching modules are also discussed. The educational value of visualizing real three-dimensional structures, and of handling real experimental results, is stressed throughout.

Bond lengths in organic and metal-organic compounds revisited: X-H bond lengths from neutron diffraction data.

The number of structures in the Cambridge Structural Database (CSD) has increased by an order of magnitude since the preparation of two major compilations of standard bond lengths in mid-1985. It is now of interest to examine whether this huge increase in data availability has implications for the mean bond-length values published in the late 1980s. Those compilations reported mean X-H bond lengths derived from rather sparse information and for rather few chemical environments. During the intervening years, the number of neutron studies has also increased, although only by a factor of around 2.25, permitting a new analysis of X-H bond-length distributions for (a) organic X = C, N, O, B, and (b) a variety of terminal and homometallic bridging transition metal hydrides. New mean values are reported here and are compared with earlier results. These new overall means are also complemented by an analysis of X-H distances at lower temperatures (T < or = 140 K), which indicates the general level of librational effects in X-H systems. The study also extends the range of chemical environments for which statistically acceptable mean X-H bond lengths can be obtained, although values from individual structures are also collated to further extend the chemical range of this compilation. Updated default 'neutron-normalization' distances for use in hydrogen-bond and deformation-density studies are also proposed for C-H, N-H and O-H, and the low-temperature analysis provides specific values for certain chemical environments and hybridization states of X.

Universal prediction of intramolecular hydrogen bonds in organic crystals.

A complete exploration of intramolecular hydrogen bonds (IHBs) has been undertaken using a combination of statistical analyses of the Cambridge Structural Database and computation of ab initio interaction energies for prototypical hydrogen-bonded fragments. Notable correlations have been revealed between computed energies, hydrogen-bond geometries, donor and acceptor chemistry, and frequencies of occurrence. Significantly, we find that 95% of all observed IHBs correspond to the five-, six- or seven-membered rings. Our method to predict a propensity for hydrogen-bond occurrence in a crystal has been adapted for such IHBs, applying topological and chemical descriptors derived from our findings. In contrast to intermolecular hydrogen bonding, it is found that IHBs can be predicted across the complete chemical landscape from a single optimized probability model, which is presented. Predictivity of 85% has been obtained for generic organic structures, which can exceed 90% for discrete classes of IHB.

Persistent hydrogen bonding in polymorphic crystal structures.

The significance of hydrogen bonding and its variability in polymorphic crystal structures is explored using new automated structural analysis methods. The concept of a chemically equivalent hydrogen bond is defined, which may be identified in pairs of structures, revealing those types of bonds that may persist, or not, in moving from one polymorphic form to another. Their frequency and nature are investigated in 882 polymorphic structures from the Cambridge Structural Database. A new method to compare conformations of equivalent molecules is introduced and applied to derive distinct subsets of conformational and packing polymorphs. The roles of chemical functionality and hydrogen-bond geometry in persistent interactions are systematically explored. Detailed structural comparisons reveal a large majority of persistent hydrogen bonds that are energetically crucial to structural stability.

Orthogonal dipolar interactions between amide carbonyl groups.

Orthogonal dipolar interactions between amide C=O bond dipoles are commonly found in crystal structures of small molecules, proteins, and protein-ligand complexes. We herein present the experimental quantification of such interactions by employing a model system based on a molecular torsion balance. Application of a thermodynamic double-mutant cycle allows for the determination of the incremental energetic contributions attributed to the dipolar contact between 2 amide C=O groups. The stabilizing free interaction enthalpies in various apolar and polar solvents amount to -2.73 kJ mol(-1) and lie in the same range as aromatic-aromatic C-H...pi and pi-pi interactions. High-level intermolecular perturbation theory (IMPT) calculations on an orthogonal acetamide/N-acetylpyrrole complex in the gas phase at optimized contact distance predict a favorable interaction energy of -9.71 kJ mol(-1). The attractive dipolar contacts reported herein provide a promising tool for small-molecule crystal design and the enhancement of ligand-protein interactions during lead optimization in medicinal chemistry.

Interaction geometries and energies of hydrogen bonds to C[double bond]O and C[double bond]S acceptors: a comparative study.

The occurrence, geometries and energies of hydrogen bonds from N-H and O-H donors to the S acceptors of thiourea derivatives, thioamides and thiones are compared with data for their O analogues - ureas, amides and ketones. Geometrical data derived from the Cambridge Structural Database indicate that hydrogen bonds to the C[double bond]S acceptors are much weaker than those to their C[double bond]O counterparts: van der Waals normalized hydrogen bonds to O are shorter than those to S by approximately 0.25 A. Further, the directionality of the approach of the hydrogen bond with respect to S, defined by the C[double bond]S...H angle, is in the range 102-109 degrees , much lower than the analogous C[double bond]O...H angle which lies in the range 127-140 degrees . Ab initio calculations using intermolecular perturbation theory show good agreement with the experimental results: the differences in hydrogen-bond directionality are closely reproduced, and the interaction energies of hydrogen bonds to S are consistently weaker than those to O, by approximately 12 kJ mol(-1), for each of the three compound classes. There are no CSD examples of hydrogen bonds to aliphatic thiones, (Csp(3))(2)C=S, consistent with the near-equality of the electronegativities of C and S. Thioureas and thioamides have electron-rich N substituents replacing the Csp(3) atoms. Electron delocalization involving C[double bond]S and the N lone pairs then induces a significant >C(delta+)[double bond]S(delta-) dipole, which enables the formation of the medium-strength C[double bond]S...H bonds observed in thioureas and thioamides.

Conformational variability of molecules in different crystal environments: a database study.

A methodology is described for analysing the Cambridge Structural Database (CSD) in terms of molecular conformations. Molecular species that have more than a single occurrence across the complete CSD are identified, either as the sole crystal component or co-crystallized with other components. Cluster analysis, based on a root-mean-square fit of coordinates and chemical connectivity, is performed to identify conformational variance for each molecule. Results are analysed in terms of the number of discrete conformations observed versus the number of crystal environments and number of acyclic torsion angles in the molecule. Special subsets of environments are also analysed, namely polymorphs, co-crystals and solvates. In general, conformational diversity increases with an increasing number of different crystal environments and with an increasing number of flexible torsion angles. Overall, molecules with one or more acyclic flexible torsion angle are observed to exist in more than one conformation in ca 40% of cases. There is evidence that solvated molecules exhibit more conformational flexibility on average, compared with polymorphs and co-crystals.

Dipolar C[triple-bond]N...C[triple-bond]N interactions in organic crystal structures: database analysis and calculation of interaction energies.

The Cambridge Structural Database (CSD) has been used to study nonbonded interactions between dipolar cyano groups. The analysis shows that C[triple-bond]N...C[triple-bond]N interactions form in an analogous manner to those involving carbonyl groups, and with the same interaction motifs: a dominant antiparallel dimer (57.5%) together with smaller populations of perpendicular (19.4%) and sheared parallel (23.0%) motifs. Ab initio calculations using intermolecular perturbation theory (IMPT) show an attractive C[triple-bond]N...C[triple-bond]N interaction in the dominant antiparallel dimer, with E(t) = -20.0 kJ mol(-1) at d(C...N) = 3.30 A and with the motif having a shear angle close to 102 degrees . The antiparallel C[triple-bond]N...C[triple-bond]N interaction is therefore slightly weaker than the analogous C=O...C=O dimer (-23.5 kJ mol(-1)), but both interactions have attractive energies similar to that of a medium-strength hydrogen bond and, where sterically favoured, they are important in the stabilization of extended crystal structures.

Knowledge-based model of hydrogen-bonding propensity in organic crystals.

A new method is presented to predict which donors and acceptors form hydrogen bonds in a crystal structure, based on the statistical analysis of hydrogen bonds in the Cambridge Structural Database (CSD). The method is named the logit hydrogen-bonding propensity (LHP) model. The approach has a potential application in identifying both likely and unusual hydrogen bonding, which can help to rationalize stable and metastable crystalline forms, of relevance to drug development in the pharmaceutical industry. Whilst polymorph prediction techniques are widely used, the LHP model is knowledge-based and is not restricted by the computational issues of polymorph prediction, and as such may form a valuable precursor to polymorph screening. Model construction applies logistic regression, using training data obtained with a new survey method based on the CSD system. The survey categorizes the hydrogen bonds and extracts model parameter values using descriptive structural and chemical properties from three-dimensional organic crystal structures. LHP predictions from a fitted model are made using two-dimensional observables alone. In the initial cases analysed, the model is highly accurate, achieving approximately 90% correct classification of both observed hydrogen bonds and non-interacting donor-acceptor pairs. Extensive statistical validation shows the LHP model to be robust across a range of small-molecule organic crystal structures.