Lorenz C. Blum

970 Million Druglike Small Molecules for Virtual Screening in the Chemical Universe Database GDB-13

GDB-13 enumerates small organic molecules containing up to 13 atoms of C, N, O, S, and Cl following simple chemical stability and synthetic feasibility rules. With 977,468,314 structures, GDB-13 is the largest publicly available small organic molecule database to date.

Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17

Drug molecules consist of a few tens of atoms connected by covalent bonds. How many such molecules are possible in total and what is their structure? This question is of pressing interest in medicinal chemistry to help solve the problems of drug potency, selectivity, and toxicity and reduce attrition rates by pointing to new molecular series. To better define the unknown chemical space, we have enumerated 166.4 billion molecules of up to 17 atoms of C, N, O, S, and halogens forming the chemical universe database GDB-17, covering a size range containing many drugs and typical for lead compounds. GDB-17 contains millions of isomers of known drugs, including analogs with high shape similarity to the parent drug. Compared to known molecules in PubChem, GDB-17 molecules are much richer in nonaromatic heterocycles, quaternary centers, and stereoisomers, densely populate the third dimension in shape space, and represent many more scaffold types.

Rapid mass spectrometric conversion of tissue biopsy samples into permanent quantitative digital proteome maps

Clinical specimens are each inherently unique, limited and nonrenewable. Small samples such as tissue biopsies are often completely consumed after a limited number of analyses. Here we present a method that enables fast and reproducible conversion of a small amount of tissue (approximating the quantity obtained by a biopsy) into a single, permanent digital file representing the mass spectrometry (MS)-measurable proteome of the sample. The method combines pressure cycling technology (PCT) and sequential window acquisition of all theoretical fragment ion spectra (SWATH)-MS. The resulting proteome maps can be analyzed, re-analyzed, compared and mined in silico to detect and quantify specific proteins across multiple samples. We used this method to process and convert 18 biopsy samples from nine patients with renal cell carcinoma into SWATH-MS fragment ion maps. From these proteome maps we detected and quantified more than 2,000 proteins with a high degree of reproducibility across all samples. The measured proteins clearly distinguished tumorous kidney tissues from healthy tissues and differentiated distinct histomorphological kidney cancer subtypes.

Quantitative variability of 342 plasma proteins in a human twin population

The degree and the origins of quantitative variability of most human plasma proteins are largely unknown. Because the twin study design provides a natural opportunity to estimate the relative contribution of heritability and environment to different traits in human population, we applied here the highly accurate and reproducible SWATH mass spectrometry technique to quantify 1,904 peptides defining 342 unique plasma proteins in 232 plasma samples collected longitudinally from pairs of monozygotic and dizygotic twins at intervals of 2–7 years, and proportioned the observed total quantitative variability to its root causes, genes, and environmental and longitudinal factors. The data indicate that different proteins show vastly different patterns of abundance variability among humans and that genetic control and longitudinal variation affect protein levels and biological processes to different degrees. The data further strongly suggest that the plasma concentrations of clinical biomarkers need to be calibrated against genetic and temporal factors. Moreover, we identified 13 cis‐SNPs significantly influencing the level of specific plasma proteins. These results therefore have immediate implications for the effective design of blood‐based biomarker studies.

Classification of organic molecules by molecular quantum numbers.

The periodic table classifies elements by increasing atomic number in periods following the principal quantum number, and allows their physicochemical properties to be rationalized. Herein, we propose a related system for organic molecules based on 42 molecular quantum numbers (MQNs), defined here as counts for simple structural features such as atom, bond and ring types, creating a multidimensional grid called MQN space. In analogy to the elements and their isotopes grouped in each entry of the periodic table, MQN isomers have identical MQNs and occupy the same position in MQN space. The MQN system is able to analyze large molecular databases and clusters compounds with similar structure, physicochemical properties and bioactivities, as illustrated for the databases ZINC and GDB-11. Organic molecules can be named by systematic nomenclature, or coded in line notations such as SMILES or InChI. These methods achieve an exact description of the molecular structure, but only provide a unidimensional classification, which is of limited use for analyzing molecular diversity. More recently, chemical space has emerged as a concept to classify large molecular databases. Chemical space is most often represented as a property space whose dimensions measure a combination of structural parameters and predicted physicochemical properties, allowing useful data mining, such as the quest for natural products analogues. While many descriptors of molecular structures and properties of varying complexity are known and may be used for defining chemical space, we set out to test whether a system based only on counts for simple structural features (MQNs) might produce an easily accessible and logical classification system for organic molecules. MQNs were considered counting atoms and bonds, polarity and topology (Table 1). MQNs were determined for the ZINC database, listing 8.4 million organic molecules, and the GDB11 database, listing 26.4 million possible molecules with up to 11 atoms of C, N, O, F, giving a total of 6 501 005 MQN combinations, or MQN bins (Table 2). Molecules with identical MQNs (MQN isomers) were strongly related structural isomers (Figure 1). Principal component analysis (PCA) showed that MQN space organizes molecules by structural types. For ZINC, 73 % of the variability is visible in the PC1/PC2 plane. PC1 mostly represents molecular size (Figure 2 a, and figure S1 in the Supporting Information). Molecules appear in elongated clusters distributed along the ascending diagonal with increasing number of rings (Figure 2 b). The number of rotatable bonds (rbc) representing molecular flexibility, the number of H-bond acceptor sites (hbam, counting nonbonding electron pairs on Nand O-atoms) and the topological surface area (TPSA in ) indicative of polarity, all increase along the descending diagonal (Figure 2 c, d & e). The calculated water–octanol partition coefficient (clog P) follows molecular size and, in part, rings and hbam (Figure 2 f). PCA of GDB-11 shows similar patterns in the PC1/PC2 plane, containing 63 % of the variability (Supporting Information, figure S2/ S3). Interestingly, compounds with similar bioactivities form groups in MQN space. Ranking by MQN distance (calculated as [a] Dr. K. T. Nguyen, L. C. Blum, R. van Deursen, Prof. Dr. J.-L. Reymond Departement of Chemistry and Biochemistry, University of Berne Freiestrasse 3, 3012 Berne (Switzerland) Fax: (+ 41) 31-631-8057 E-mail : jean-louis.reymond@ioc.unibe.ch Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cmdc.200900317. Table 1. Molecular quantum numbers.

A Searchable Map of PubChem

The database PubChem was classified using 42 integer value descriptors of molecular structure, here called molecular quantum numbers (MQNs), which count atoms and bond types, polar groups, and topological features. Principal component analysis of the MQN data set shows that PubChem compounds occupy a partially filled elliptical cone in the (PC1,PC2,PC3) space whose axis is the first principal component PC1 (65% variability) representing molecular size, and the ellipse axes are PC2 (18% variability, representing structural flexibility) and PC3 (7% variability, representing polarity). A visual overview of PubChem is provided by color-coded representations of the (PC2,PC3) plane. The MQNs form a scalar fingerprint which can be used to measure the similarity between pairs of molecules and enable ligand-based virtual screening, as illustrated for the enrichment of bioactives from the DUD data set from PubChem. An MQN-annotated version of PubChem with an MQN-similarity search tool is available at www.gdb.unibe.ch .

The enumeration of chemical space

In the field of medicinal chemistry, the chemical space describes the ensemble of all organic molecules to be considered when searching for new drugs (estimated >1060 molecules), as well as the property spaces in which these molecules are placed for the sake of describing them. Molecules can be enumerated computationally by the millions, which was first undertaken in the field of computer‐aided structure elucidation. Scoring the enumerated virtual libraries by virtual screening has recently become an attractive strategy to prioritize compounds for synthesis and testing. Enumeration methods include combinatorial linking of fragments, genetic algorithms based on cycles of enumeration and selection by ligand‐based or target‐based scoring functions, and exhaustive enumeration from first principles. The chemical space of molecules following simple rules of chemical stability and synthetic feasibility has been enumerated up to 13 atoms of C, N, O, Cl, S, forming the GDB‐13 database with 977 million structures. The database has been organized in a 42‐dimensional chemical space using molecular quantum numbers (MQN) as descriptors, which can be visualized by projection in two dimensions by principal component analysis, and searched within seconds using a Web browser available at www.gdb.unibe.ch.

Identification of selective norbornane-type aspartate analogue inhibitors of the glutamate transporter 1 (GLT-1) from the chemical universe generated database (GDB)

A variety of conformationally constrained aspartate and glutamate analogues inhibit the glutamate transporter 1 (GLT-1, also known as EAAT2). To expand the search for such analogues, a virtual library of aliphatic aspartate and glutamate analogues was generated starting from the chemical universe database GDB-11, which contains 26.4 million possible molecules up to 11 atoms of C, N, O, F, resulting in 101026 aspartate analogues and 151285 glutamate analogues. Virtual screening was realized by high-throughput docking to the glutamate binding site of the glutamate transporter homologue from Pyrococcus horikoshii (PDB code: 1XFH ) using Autodock. Norbornane-type aspartate analogues were selected from the top-scoring virtual hits and synthesized. Testing and optimization led to the identification of (1R*,2R*,3S*,4R*,6R*)-2-amino-6-phenethyl-bicyclo[2.2.1]heptane-2,3-dicarboxylic acid as a new inhibitor of GLT-1 with IC(50) = 1.4 μM against GLT-1 and no inhibition of the related transporter EAAC1. The systematic diversification of known ligands by enumeration with help of GDB followed by virtual screening, synthesis, and testing as exemplified here provides a general strategy for drug discovery.

Discovery of α7-Nicotinic Receptor Ligands by Virtual Screening of the Chemical Universe Database GDB-13

The chemical universe database GDB-13 enumerates 977 million organic molecules up to 13 atoms of C, N, O, Cl, and S that are virtually possible following simple rules for chemical stability and synthetic feasibility. Analogs of nicotine were identified in GDB-13 using the city-block distance in MQN-space (CBD(MQN)) as a similarity measure, combined with a restriction eliminating problematic structural elements. The search was carried out with a Web browser available at www.gdb.unibe.ch . This virtual screening procedure selected 31 504 analogs of nicotine from GDB-13, from which 48 were known nicotinic ligands reported in Chembl. An additional 60 virtual screening hits were purchased and tested for modulation of the acetylcholine signal at the human α7 nAChR expressed in Xenopus oocytes, which led to the identification of three previously unknown inhibitors. These experiments demonstrate for the first time the use of GDB-13 for ligand discovery.

Exploring the Chemical Space of Known and Unknown Organic Small Molecules at www.gdb.unibe.ch

Organic small molecules are of particular interest for medicinal chemistry since they comprise many biologically active compounds which are potential drugs. To understand this vast chemical space, we are enumerating all possible organic molecules to create the chemical universe database GDB, which currently comprises 977 million molecules up to 13 atoms of C, N, O, Cl and S. Furthermore, we have established a simple classification method for organic molecules in form of the MQN (molecular quantum numbers) system, which is an equivalent of the periodic system of the elements. Despite its simplicity the 42 dimensional MQN system is surprisingly relevant with respect to bioactivity, as evidenced by the fact that groups of biosimilar compounds form close groups in MQN space. The MQN space of the known organic molecules in PubChem and of the unknown molecules in the Chemical Universe Database GDB-13 can be searched interactively using browser tools freely accessible at www.gdb.unibe.ch.