Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Pliability”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24Linked to original sources
PubMed2001

Hydrophobic core malleability of a de novo designed three-helix bundle protein.

De novo protein design provides a tool for testing the principles that stabilize the structures of proteins. Recently, we described the design and structure determination of alpha(3)D, a three-helix bundle protein with a well-packed hydrophobic core. Here, we test the malleability and adaptability of this protein's structure by mutating a small, Ala residue (A60) in its core to larger, hydrophobic side-chains, Leu and Ile. Such changes introduce strain into the structures of natural proteins, and therefore generally destabilize the native state. By contrast, these mutations were slightly stabilizing ( approximately 1.5 kcal mol(-1)) to the tertiary structure of alpha(3)D. The value of DeltaC(p) for unfolding of these mutants was not greatly affected relative to wild-type, indicating that the change in solvent accessibility for unfolding was similar. However, two-dimensional heteronuclear single quantum coherence spectra indicate that the protein adjusts to the introduction of steric bulk in different ways. A60L-alpha(3)D showed serious erosion in the dispersion of both the amide backbone as well as the side-chain methyl chemical shifts. By contrast, A60I-alpha(3)D showed excellent dispersion of the backbone resonances, and selective changes in dispersion of the aliphatic side-chains proximal to the site of mutation. Together, these data suggest that alpha(3)D, although folded into a unique three-dimensional structure, is nevertheless more malleable and flexible than most natural, native proteins.

Amino Acid Substitution↗
PubMed2000

Crystal structure of human GM2-activator protein with a novel beta-cup topology.

GM2 activator protein (GM2-AP) belongs to a small group of non- enzymatic lysosomal proteins that act as cofactors in the sequential degradation of gangliosides. It has been postulated that GM2-AP extracts single GM2 molecules from membranes and presents them in soluble form to beta-hexosaminidase A for cleavage of N-acetyl-d-galactosamine and conversion to GM3. The high affinity of GM2-AP for GM2 is based on specfic recognition of the oligosaccharide moiety as well as the ceramide lipid tail. Genetic defects in GM2-AP result in an atypical form of Tay-Sachs disease known as variant AB GM2 gangliosidosis. The 2.0 A resolution crystal structure of GM2-AP reported here reveals a previously unobserved fold whose main feature is an eight-stranded cup-shaped anti-parallel beta-pleated sheet. The striking feature of the GM2-AP structure is that it possesses an accessible central hydrophobic cavity rather than a buried hydrophobic core. The dimensions of this cavity (12 Ax14 Ax22 A) are suitable for binding 18-carbon lipid acyl chains. Flexible surface loops and a short alpha-helix decorate the mouth of the beta-cup and may control lipid entry to the cavity.

Amino Acid Sequence↗
PubMed2000

Identification of the phospholipid-binding site of human beta(2)-glycoprotein I domain V by heteronuclear magnetic resonance.

To understand the mechanism of the interaction between human beta(2)-glycoprotein I (beta(2)-GPI) and negatively charged phospholipids, we determined the three-dimensional solution structure of the fifth domain of beta(2)-GPI by heteronuclear multidimensional NMR. The results showed that the molecule is composed of well-defined four anti-parallel beta-strands and two short alpha-helices, as well as a long highly flexible loop. Backbone dynamic analysis demonstrated significant mobility of the flexible loop on a subnanosecond time scale. Structural modeling of the nicked fifth domain, in which the Lys317-Thr318 peptide bond was specifically cleaved, revealed the importance of this long C-terminal loop for the interaction between beta(2)-GPI and negatively charged phospholipids. A titration experiment with the anionic surfactant SDS showed that this highly mobile loop, as well as the short beta-hairpin between betaC and betaD strands, which is rich in positively charged residues, specifically interact with the surfactant. The mobile loop, together with the surrounding positively charged residues, probably construct the binding site for negatively charged phospholipids such as cardiolipin.

Amino Acid Sequence↗
PubMed2001

Structure-activity relationships in flexible protein domains: regulation of rho GTPases by RhoGDI and D4 GDI.

The guanine dissociation inhibitors RhoGDI and D4GDI inhibit guanosine 5'-diphosphate dissociation from Rho GTPases, keeping these small GTPases in an inactive state. The GDIs are made up of two domains: a flexible N-terminal domain of about 70 amino acid residues and a folded 134-residue C-terminal domain. Here, we characterize the conformation of the N-terminal regions of both RhoGDI and D4GDI using a series of NMR experiments which include (15)N relaxation and amide solvent accessibility measurements. In each protein, two regions with tendencies to form helices are identified: residues 36 to 58 and 9 to 20 in RhoGDI, and residues 36 to 57 and 20 to 25 in D4GDI. To examine the functional roles of the N-terminal domain of RhoGDI, in vitro and in vivo functional assays have been carried out with N-terminally truncated proteins. These studies show that the first 30 amino acid residues are not required for inhibition of GDP dissociation but appear to be important for GTP hydrolysis, whilst removal of the first 41 residues completely abolish the ability of RhoGDI to inhibit GDP dissociation. The combination of structural and functional studies allows us to explain why RhoGDI and D4GDI are able to interact in similar ways with the guanosine 5'-diphosphate-bound GTPase, but differ in their ability to regulate GTP-bound forms; these functional differences are attributed to the conformational differences of the N-terminal domains of the guanosine 5'-diphosphate dissociation inhibitors. Therefore, the two transient helices, appear to be associated with different biological effects of RhoGDI, providing a clear example of structure-activity relationships in a flexible protein domain.

Amides↗
PubMed2001

Structure and dynamics of micelle-bound neuropeptide Y: comparison with unligated NPY and implications for receptor selection.

The biological importance of the neuropeptide Y (NPY) has steered a number of investigations about its solution structure over the last 20 years. Here, we focus on the comparison of the structure and dynamics of NPY free in solution to when bound to a membrane mimetic, dodecylphosphocholine (DPC) micelles, as studied by 2D (1)H NMR spectroscopy. Both, free in solution and in the micelle-bound form, the N-terminal segment (Tyr1-Glu15) is shown to extend like a flexible tail in solution. This is not compatible with the PP-fold model for NPY that postulates backfolding of the flexible N terminus onto the C-terminal helix. The correlation time (tau(c)) of NPY in aqueous solution, 5.5 (+/-1.0) ns at 32 degrees C, is only consistent with its existence in a dimeric form. Exchange contributions especially enhancing transverse relaxation rates (R(2)) of residues located on one side of the C-terminal helix of the molecule are supposed to originate from dimerization of the NPY molecule. The dimerization interface was directly probed by looking at (15)N-labeled NPY/spin-labeled [TOAC34]-[(14)N]-NPY heterodimers and revealed both parallel and anti-parallel alignment of the helices. The NMR-derived three-dimensional structure of micelle-bound NPY at 37 degrees C and pH 6.0 is similar but not identical to that free in solution. The final set of 17 lowest-energy DYANA structures is particularly well defined in the region of residues 21-31, with a mean pairwise RMSD of 0.23 A for the backbone heavy atoms and 0.85 A for all heavy atoms. The combination of NMR relaxation data and CD measurements clearly demonstrates that the alpha-helical region Ala18-Thr32 is more stable, and the C-terminal tetrapeptide becomes structured only in the presence of the phosphocholine micelles. The position of NPY relative to the DPC micelle surface was probed by adding micelle integrating spin labels. Together with information from (1)H,(2)H exchange rates, we conclude that the interaction of NPY with the micelle is promoted by the amphiphilic alpha-helical segment of residues Tyr21-Thr32. NPY is located at the lipid-water interface with its C-terminal helix parallel to the membrane surface and penetrates the hydrophobic interior only via insertions of a few long aliphatic or aromatic side-chains. From these data we can demonstrate that the dimer interface of neuropeptide Y is similar to the interface of the monomer binding to DPC-micelles. We speculate that binding of the NPY monomer to the membrane is an essential key step preceeding receptor binding, thereby pre-orientating the C-terminal tetrapeptide and possibly inducing the bio-active conformation.

Amino Acid Sequence↗
PubMed2000

From minichaperone to GroEL 1: information on GroEL-polypeptide interactions from crystal packing of minichaperones.

We are reconstructing the mechanism of action of GroEL by a reductionist approach of isolating its minimal fragment that has residual activity (the "minichaperone" core) and then identifying how additional elements of structure confer further activity and function. We report here the 2.0 A resolution crystal structure of the minichaperone GroEL(193-345). The structure provides further clues on the nature of GroEL-polypeptide substrate interactions, because two molecules in the asymmetric unit interact by the binding of one molecule in the active site of its partner, thus mimicking a chaperone-polypeptide substrate complex. The results may explain some experimental observations, including the preference of GroEL for net positive charges (mediated by Glu238 and Glu257) and the key role of Tyr203 in mediating polypeptide binding. The larger binding site identified by these studies forms a continuous surface near the opening of the central cavity of GroEL that can accommodate a wide range of non-native protein conformations that differ in size and in structural and chemical properties.

Binding Sites↗
PubMed2001

The DNA-binding domain in the Bacillus subtilis transition-state regulator AbrB employs significant motion for promiscuous DNA recognition.

AbrB is a Bacillus subtilis protein responsible for regulating a diverse array of unrelated genes during periods of sub-optimal growth conditions. DNA binding by AbrB is unique in that sequence recognition is specific, yet no obvious consensus sequence of bound promoter regions is apparent. The N-terminal domain is a recently characterized representative of a novel class of DNA-binding proteins that possess a looped-hinge helix DNA-binding topology. Although the structural characterization of this DNA-binding topology contributed to an understanding of the architectural basis for recognition of DNA target sequences, specific mechanisms responsible for promiscuity in DNA sequence recognition still were not apparent. Analysis of (15)N backbone relaxation parameters shows that dynamic motion of regions directly linked to DNA binding show concerted motion on the microsecond-millisecond timescale. Furthermore, dynamic motion of the hinge region suggests that the DNA-binding region is capable of conformational orientations that allow it to accommodate DNA sequence variability in the cognate binding sites.

Amino Acid Sequence↗
PubMed2001

Crystal structure of soybean proglycinin A1aB1b homotrimer.

Soybean glycinin is a member of the 11 S globulin family. The crystal structure of proglycinin was determined by X-ray crystallography at 2.8 A resolution with an R-factor of 0.199 and a free R-factor of 0.250. A trimer molecule was found in an asymmetric unit of crystals. The trimer model contains three A1aB1b subunits and comprises 1128 amino acid residues and 34 water molecules. The constituent protomers of the homo-trimeric protein are arranged around a 3-fold symmetry axis with dimensions of 95 Ax95 Ax40 A. The protomer model is composed of five fragments which correspond roughly to conserved regions based on the sequence alignment of various 11 S globulins. The core of the protomer consists of two jelly-roll beta-barrels and two extended helix domains. This structure of proglycinin is similar to those of canavalin and phaseolin belonging to the 7 S globulin family, strongly supporting the hypothesis that both 7 S and 11 S globulins are derived from a common ancestor. The inter and intra-chain disulfide bonds conserved in the 11 S globulin family are clearly observed. It is found that the face with the inter-chain disulfide bond (IE face) contains more hydrophobic residues than that with the intra-chain disulfide bond. This suggests that a mature hexamer is formed by the interaction between the IE faces after processing.

Amino Acid Sequence↗
PubMed2001

Characterization of two hydrophobic methyl clusters in HIV-1 protease by NMR spin relaxation in solution.

Nearly 50 % of the amino acid residues of HIV-1 protease contain methyl side-chains, most of which appear to be organized into two clusters: the inner cluster that nearly surrounds the active site and the outer cluster that contains the hydrophobic core which stabilizes the inhibitor-free protease structure. NMR relaxation experiments sensitive to motions of methyl groups on the sub-nanosecond and the milli-microsecond time-scales revealed flexible methyl groups in residues that link the two clusters, the methyl groups of L10, L23, V75, and L76. We hypothesize that flexibility at the junctions of these clusters allows the protease to minimize conformational changes upon drug-binding. The two-methyl cluster motif appears to be a common structural feature among retroviral proteases and may play a similar role throughout this family of enzymes.

Amino Acid Substitution↗
PubMed2001

A 2.1 A resolution structure of an uncleaved alpha(1)-antitrypsin shows variability of the reactive center and other loops.

Serpin (serine protease inhibitor) proteins are involved in diverse physiological processes including inflammation, coagulation, matrix remodeling, and cell differentiation. Deficiency of normal serpin functions leads to various hereditary diseases. Besides their clinical importance, serpin proteins draw much attention due to the large conformational changes that occur upon interaction with proteases. We present here the crystal structure of an uncleaved alpha(1)-antitrypsin determined by the multiple isomorphous replacement method and refined to 2.1 A resolution. The structure, which is the first active serpin structure based on experimental phases, reveals novel conformations in the flexible loops, including the proximal hinge region of the reactive center loop and the surface cavity region in the central beta-sheet, sheet A. The determined loop conformation explains the results of recent mutagenesis studies and provides detailed insights into the protease inhibition mechanism. The high-resolution structure of active alpha(1)-antitrypsin also provides evidence for the existence of localized van-der-Waals strain in the central hydrophobic core.

Amino Acid Sequence↗
PubMed2001

Conformation and rigidity of DNA microcircles containing waf1 response element for p53 regulatory protein.

The tumor-suppressor activity of p53 is closely related to its DNA-binding properties. It binds a number of DNA response-elements and it is likely that these share a common structural feature. Here, we present a new, general method to determine the absolute twist of flexible DNA promoter sequences based on direct imaging of the topology of microcircles containing the sequences. We have used magnetically driven dynamic force microscopy ("MacMode" AFM) to observe, in solution, the conformation of 168 base-pair DNA microcircles, each containing four equally spaced copies of the waf1/cip1/p21 p53 response-element. Analysis of the images showed that the microcircles are markedly puckered with a small excess of negatively writhed molecules. The average measured values of writhe are 0.109+/-0.013 (for 60 positively writhed molecules) and -0.098+/-0.011 (for 65 negatively writhed molecules). These values lead directly to a difference in linking number for the positively and negatively writhed molecules prior to ligation, from which we derive a twist mismatch of 178 degrees (overtwist). This is 44.5 degrees for each 42-mer precursor containing a single waf1/cip1/p21 p53 response-element, in good agreement with the range of values deduced by indirect biochemical techniques. The two values of writhe may also be used to determine the ratio of the bending (B) to twisting (C) rigidity, yielding B/C=0.23. This is about one-third of the value for long, random-sequence DNA, suggesting that the waf1/cip1/p21 p53 response-element is extremely flexible, a result that is also consistent with indirect biochemical experiments. These results support the idea, proposed by us earlier, that torsional stress may play a role in the regulation of p53 binding through modulation of twist at the binding site.

Aluminum Silicates↗
PubMed2001

Structural basis of diverse sequence-dependent target recognition by the 8 kDa dynein light chain.

Dyneins are multi-subunit molecular motors that translocate molecular cargoes along microtubules. Other than acting as an essential component of the dynein motor complex, the 89-residue subunit of dynein light chain (DLC8) also regulates a number of other biological events by binding to various proteins and enzymes. Currently known DLC8 targets include neuronal nitric oxide synthase; the proapoptotic Bcl-2 family member protein designated Bim; a Drosophila RNA localization protein Swallow, myosin V, neuronal scaffolding protein GKAP, and IkappaBalpha, an inhibitor of the NFkappaB transcription factor. The DLC8-binding domains of the various targets are confined within a short, continuous stretch of amino acid residues. However, these domains do not share any obvious sequence homology with each other. Here, the three-dimensional structures of DLC8 complexed with two peptides corresponding to the DLC8-binding domains of neuronal nitric oxide synthase and Bim, respectively, were determined by NMR spectroscopy. Although the two DLC8-binding peptides have entirely different amino acid sequences, both peptides bind to the protein with a remarkable similar conformation by engaging the symmetric DLC8 dimer through antiparallel beta-sheet augmentation via the beta2 strand of the protein. Structural comparison indicates that the two target peptides use different regions within the conformational flexible peptide-binding channels to achieve binding specificity. We have also re-determined the apo-form solution structure of DLC8 in this work. The structures of the DLC8/target peptide complexes, together with the dynamic properties of the protein, provide a molecular basis of DLC8's diverse amino acid sequence-dependent target recognition.

Amino Acid Sequence↗
PubMed2001

Solution structure of Grb2 reveals extensive flexibility necessary for target recognition.

Grb2 is an adaptor protein composed of a single SH2 domain flanked by two SH3 domains. Grb2 functions as an important evolutionary conserved link between a variety of cell membrane receptors and the Ras/MAP kinase-signaling cascade. Here, we describe the solution structure of Grb2 as revealed by NMR and small angle X-ray scattering measurements. We demonstrate that Grb2 is a flexible protein in which the C-terminal SH3 domain is connected to the SH2 domain via a flexible linker. This is in contrast to the previously described Grb2 crystal structure, which showed a compact structure with intramolecular contact between two SH3 domains. Binding experiments on Grb2 and peptides containing two different proline-rich sequences indicate that Grb2 adapts the relative position and orientation of the two SH3 domains to bind bivalently to the target peptide sequences.

Adaptor Proteins, Signal Transducing↗
PubMed2001

Statistical theory for protein combinatorial libraries. Packing interactions, backbone flexibility, and the sequence variability of a main-chain structure.

Combinatorial experiments provide new ways to probe the determinants of protein folding and to identify novel folding amino acid sequences. These types of experiments, however, are complicated both by enormous conformational complexity and by large numbers of possible sequences. Therefore, a quantitative computational theory would be helpful in designing and interpreting these types of experiment. Here, we present and apply a statistically based, computational approach for identifying the properties of sequences compatible with a given main-chain structure. Protein side-chain conformations are included in an atom-based fashion. Calculations are performed for a variety of similar backbone structures to identify sequence properties that are robust with respect to minor changes in main-chain structure. Rather than specific sequences, the method yields the likelihood of each of the amino acids at preselected positions in a given protein structure. The theory may be used to quantify the characteristics of sequence space for a chosen structure without explicitly tabulating sequences. To account for hydrophobic effects, we introduce an environmental energy that it is consistent with other simple hydrophobicity scales and show that it is effective for side-chain modeling. We apply the method to calculate the identity probabilities of selected positions of the immunoglobulin light chain-binding domain of protein L, for which many variant folding sequences are available. The calculations compare favorably with the experimentally observed identity probabilities.

Amino Acid Sequence↗
PubMed2001

Tiny TIM: a small, tetrameric, hyperthermostable triosephosphate isomerase.

Comparative structural studies on proteins derived from organisms with growth optima ranging from 15 to 100 degrees C are beginning to shed light on the mechanisms of protein thermoadaptation. One means of sustaining hyperthermostability is for proteins to exist in higher oligomeric forms than their mesophilic homologues. Triosephosphate isomerase (TIM) is one of the most studied enzymes, whose fold represents one of nature's most common protein architectures. Most TIMs are dimers of approximately 250 amino acid residues per monomer. Here, we report the 2.7 A resolution crystal structure of the extremely thermostable TIM from Pyrococcus woesei, a hyperthermophilic archaeon growing optimally at 100 degrees C, representing the first archaeal TIM structure. P. woesei TIM exists as a tetramer comprising monomers of only 228 amino acid residues. Structural comparisons with other less thermostable TIMs show that although the central beta-barrel is largely conserved, severe pruning of several helices and truncation of some loops give rise to a much more compact monomer in the small hyperthermophilic TIM. The classical TIM dimer formation is conserved in P. woesei TIM. The extreme thermostability of PwTIM appears to be achieved by the creation of a compact tetramer where two classical TIM dimers interact via an extensive hydrophobic interface. The tetramer is formed through largely hydrophobic interactions between some of the pruned helical regions. The equivalent helical regions in less thermostable dimeric TIMs represent regions of high average temperature factor. The PwTIM seems to have removed these regions of potential instability in the formation of the tetramer. This study of PwTIM provides further support for the role of higher oligomerisation states in extreme thermal stabilisation.

Amino Acid Sequence↗
PubMed2001

Docking ligands onto binding site representations derived from proteins built by homology modelling.

Due to the abundant sequence information available from genome projects, an increasing number of structurally unknown proteins, homologous to examples of known 3D structure, will be discovered as new targets for drug design. Since homology models do not provide sufficient accuracy to apply common drug design tools, a new approach, DragHome, has been developed to dock ligands into such approximate protein models. DragHome combines information from homology modelling with ligand data, used by and derived from 3D quantitative structure-activity relationships (QSAR). The binding-site of a model-built protein is analysed in terms of putative ligand interaction sites and translated via Gaussian functions into a functional binding-site description represented by physico-chemical properties. Ligands to be docked onto these binding-site representations are similarly translated into a description based on Gaussian functions. The docking is computed by optimising the overlap between the functional description of the binding site and the ligand, generating multiple solutions. For a set of different ligands, these solutions are ranked according to the internal similarity consistance among the various ligands in the binding modes obtained from docking. DragHome has been validated at examples for which crystal structures are available: structurally distinct thrombin inhibitors were docked onto models of thrombin generated from serine proteases of 28 to 40 % sequence identity, yielding ligand binding modes with an average RMS deviation of 1.4 A. Mostly the near-native solutions are ranked best. Molecular flexibility of ligands can be considered in terms of pre-calculated multiple conformers. DragHome has been used to automatically generate an alignment of 88 thrombin inhibitors, for which a significant 3D QSAR model could be derived. The contribution maps resulting from this analysis can be interpreted with respect to the surrounding protein model. They highlight inconsistencies and deficiencies present in the model. In future developments, this information could be fed back into a subsequent modelling step to improve the protein model.

Algorithms↗
PubMed2001

Unrestrained 5 ns molecular dynamics simulation of a cisplatin-DNA 1,2-GG adduct provides a rationale for the NMR features and reveals increased conformational flexibility at the platinum binding site.

A 5 ns unrestrained molecular dynamics (MD) simulation of the DNA duplex d(GCCG*G*ATCGC)-d(GCGATCCGGC), bearing a cis-Pt(NH(3))(2)(2+) unit crosslinking the two G* guanine bases, is reported. The MD trajectory was a posteriori correlated with NMR data determined for the same adduct, and it is shown that interproton distances and the characteristic chemical shifts are accounted for by the simulation. The simulation and its confrontation with the NMR data have confirmed the finding derived early from static models that the cytosine complementary to the 5' G*, C17, is mobile with respect to its adjacent bases. However, in contrast to our previous description of this mobility, which included rupture of the Watson-Crick hydrogen bonds and formation of non-Watson-Crick hydrogen bonds, the MD simulation indicated that the G*4-C17 pair moves continuously along a trajectory roughly perpendicular to the local helix axis, with retention of all three Watson-Crick hydrogen bonds. The simulation indicated the reversible formation of a hydrogen bond between the 5' oriented NH(3) ligand of platinum and the C3pG*4 phosphate group, in accord with our former prediction. Furthermore, the simulation has disclosed previously undetected BI <=> BII transitions at the G*5pA6 and A6pT7 steps, connected to formation/rupture of a hydrogen bond between the 3' oriented NH(3) ligand of platinum and the N7 atom of A6. All these conformational equilibria affect the form of the minor groove and increase the conformational flexibility at the platination site, and are thus likely to facilitate recognition by cellular proteins.

Base Sequence↗
PubMed2001

SuperStar: improved knowledge-based interaction fields for protein binding sites.

SuperStar is an empirical method for identifying interaction sites in proteins, based entirely on experimental information about non-bonded interactions occurring in small-molecule crystal structures, taken from the IsoStar database. We describe recent modifications and additions to SuperStar, validating the results on a test set of 122 X-ray structures of protein-ligand complexes. In this validation, propensity maps are generated for all the binding sites of these proteins, using four different probes: a charged NH(+)(3) nitrogen atom, a carbonyl oxygen atom, a hydroxyl oxygen atom and a methyl carbon atom. Next, the maps are compared with the experimentally observed positions of ligand atoms of these types. A peak-searching algorithm is introduced that highlights potential interaction hot spots. For the three hydrogen-bonding probes - NH(+)(3) nitrogen atom, carbonyl oxygen atom and hydroxyl oxygen atom - the average distance from the ligand atom to the nearest SuperStar peak is 1.0-1.2 A (0.8-1.0 A for solvent-inaccessible ligand atoms). For the methyl carbon atom probe, this distance is about 1.5 A, probably because interactions to methyl groups are much less directional. The most important addition to SuperStar is the enabling of propensity maps around metal centres - Ca(2+), Mg(2+) and Zn(2+) - in protein binding sites. The results are validated on a test set of 24 protein-ligand complexes that have a metal ion in their binding site. Coordination geometries are derived automatically, using only the protein atoms that coordinate to the metal ion. The correct coordination geometry is derived in approximately 75 % of the cases. If the derived geometry is assumed during the SuperStar calculation, the average distance from a ligand atom coordinating to the metal ion to the nearest peak in the propensity map for an oxygen probe is 0.87(7) A. If the correct coordination geometry is imposed, this distance reduces to 0.59(7)A. This indicates that the SuperStar predictions around metal-binding sites are at least as good as those around other protein groups. Using clustering techniques, a non-redundant set of probes is selected from the set of probes available in the IsoStar database. The performance in SuperStar of all these probes is tested on the test set of protein-ligand complexes. With the exception of the "ether oxygen" probe and the "any NH(+)" probe, all new probes perform as well as the four probes introduced first.

Algorithms↗