Computational drug repositioning: from data to therapeutics.

Traditionally, most drugs have been discovered using phenotypic or target-based screens. Subsequently, their indications are often expanded on the basis of clinical observations, providing additional benefit to patients. This review highlights computational techniques for systematic analysis of transcriptomics (Connectivity Map, CMap), side effects, and genetics (genome-wide association study, GWAS) data to generate new hypotheses for additional indications. We also discuss data domains such as electronic health records (EHRs) and phenotypic screening that we consider promising for novel computational repositioning methods.

Impact of human genome sequencing for in silico target discovery.

The year 2000 stands as a landmark in modern biology: the first draft of the human genome sequence has been completed. For the pharmaceutical industry, this achievement provides tremendous opportunities because the genomic sequence exposes all human drug targets for therapeutic intervention. The challenge for the pharmaceutical companies is to exploit this definitive resource for the identification of potential molecular targets, rapid characterization of their function and validation of their involvement in disease pathology. Bioinformatics approaches provide increasingly crucial tools to systematically support this exploratory target drug discovery activity.

Identification and characterization of a novel human vanilloid receptor-like protein, VRL-2.

Remarkable progress has been made recently in identifying a new gene family related to the capsaicin (vanilloid) receptor, VR1. Using a combination of in silico analysis of expressed sequence tag (EST) databases and conventional molecular cloning, we have isolated a novel vanilloid-like receptor, which we call VRL-2, from human kidney. The translated gene shares 46% and 43% identity with VR1 and VRL-1, respectively, and maps to chromosome 12q23-24.1, a locus associated with bipolar affective disorder. VRL-2 mRNA was most strongly expressed in the trachea, kidney, and salivary gland. An affinity-purified antibody against a peptide incorporating the COOH terminal of the receptor localized VRL-2 immunolabel in the distal tubules of the kidney, the epithelial linings of both trachea and lung airways, serous cells of submucosal glands, and mononuclear cells. Unlike VR1 and VRL-1, VRL-2 was not detected in cell bodies of dorsal root ganglia (DRG) or sensory nerve fibers. However, VRL-2 was found on sympathetic and parasympathetic nerve fibers, such as those innervating the arrector pili smooth muscle in skin, sweat glands, intestine, and blood vessels. At least four vanilloid receptor-like genes exist, the newest member, VRL-2 is found in airway and kidney epithelia and in the autonomic nervous system.

Rapid in silico cloning of genes using expressed sequence tags (ESTs).

Expressed sequence tags (ESTs) are short single-pass DNA sequences obtained from either end of cDNA clones. These ESTs are derived from a vast number of cDNA libraries obtained from different species. Human ESTs are the bulk of the data and have been widely used to identify new members of gene families, as markers on the human chromosomes, to discover polymorphism sites and to compare expression patterns in different tissues or pathologies states. Information strategies have been devised to query EST databases. Since most of the analysis is performed with a computer, the term "in silico" strategy has been coined. In this chapter we will review the current status of EST databases, the pros and cons of EST-type data and describe possible strategies to retrieve meaningful information.

An integrated map of human 6q22.3-q24 including a 3-Mb high-resolution BAC/PAC contig encompassing a QTL for fetal hemoglobin.

Genetic studies have previously assigned a quantitative trait locus (QTL) for hemoglobin F and F cells to a region of approximately 4 Mb between the markers D6S408 and D6S292 on chromosome 6q23. An initial yeast artificial chromosome contig of 13 clones spanning this region was generated. Further linkage analysis of an extended kindred refined the candidate interval to 1-2 cM, and key recombination events now place the QTL within a region of <800 kb. We describe a high-resolution bacterial clone contig spanning 3 Mb covering this critical region. The map consists of 223 bacterial artificial chromosome (BAC) and 100 P1 artificial chromosome (PAC) clones ordered by sequence-tagged site (STS) content and restriction fragment fingerprinting with a minimum tiling path of 22 BACs and 1 PAC. A total of 194 STSs map to this interval of 3 Mb, giving an average marker resolution of approximately one per 15 kb. About half of the markers were novel and were isolated in the present study, including three CA repeats and 13 single nucleotide polymorphisms. Altogether 24 expressed sequence tags, 6 of which are unique genes, have been mapped to the contig.

In silico cloning of a new protein kinase, Aik2, related to Drosophila Aurora using the new tool: EST Blast.

In this short communication we report for the first time to our knowledge the use of ESTBlast to in silico clone a new gene and a step by step description of this particular in silico cloning project.

Cloning of STK13, a third human protein kinase related to Drosophila aurora and budding yeast Ipl1 that maps on chromosome 19q13.3-ter.

This report describes the identification of a cDNA encoding STK13, a third human protein kinase related to the Drosophila Aurora and the budding yeast Ipl1 kinases. After screening of a human placental cDNA library with a Xenopus laevis cDNA encoding the pEg2 protein kinase and 5' RACE on testis mRNA, a full-length cDNA was isolated. The chromosomal localization of STK13 on 19q13.3-ter between the markers D19S210 and D19S218 was established by a combination of somatic cell and radiation hybrid panel PCR screening. The localization of STK13 on human chromosome 19 was confirmed by fluorescence in situ hybridization (FISH) using a genomic clone containing STK13 as a probe.

WEBMAP: radiation hybrid mapping on the WWW.

SUMMARY: A Java interface to radiation hybrid (RH) mapping software is described which enables users to build and interactively refine RH maps over the web. AVAILABILITY: The Java applets described here are available on the internet at http://www.oxmol.com/biolib/webmap/.

A new dynamic tool to perform assembly of expressed sequence tags (ESTs).

MOTIVATION: Expressed Sequence Tags (ESTs) are short single-pass DNA sequences obtained from either ends of cDNA clones. To exploit these sequences efficiently, a dynamic Web-tool has been developed which uses these data to perform fast virtual cloning of cDNAs. RESULTS: Starting with a query sequence, the user is able to identify related ESTs and extend the sequence of interest step by step, possibly to a full-length transcript. Graphical views of the clustering are used to monitor the progress of a particular 'cloning' project. Potential open reading frames are detected by positional base preference, and hyperlinks to other Worldwide Web sites allows the user to retrieve information relevant to each EST in a cluster (e.g. sequence traces, clone size, plate position). Apart from cDNA cloning, this tool also provides a mechanism for collating gene families and polymorphism sites.

Conservation of synteny between the genome of the pufferfish (Fugu rubripes) and the region on human chromosome 14 (14q24.3) associated with familial Alzheimer disease (AD3 locus)

The genome of the pufferfish (Fugu rubripes) (400 Mb) is approximately 7.5 times smaller than the human genome, but it has a similar gene repertoire to that of man. If regions of the two genomes exhibited conservation of gene order (i.e., were syntenic), it should be possible to reduce dramatically the effort required for identification of candidate genes in human disease loci by sequencing syntenic regions of the compact Fugu genome. We have demonstrated that three genes (dihydrolipoamide succinyltransferase, S31iii125, and S20i15), which are linked to FOS in the familial Alzheimer disease focus (AD3) on human chromosome 14, have homologues in the Fugu genome adjacent to Fugu cFOS. The relative gene order of cFOS, S31iii125, and S20i15 was the same in both genomes, but in Fugu these three genes lay within a 12.4-kb region, compared to >600 kb in the human AD3 locus. These results demonstrate the conservation of synteny between the genomes of Fugu and man and highlight the utility of this approach for sequence-based identification of genes in human disease loci.

Evolutionary dynamics of non-coding sequences within the class II region of the human MHC.

About 40% (350 kb) of the human MHC class II region has been sequenced and a coordinated effort to sequence the entire MHC is underway. In addition to the coding information (22 genes/pseudogenes), the non-coding sequences reveal novel information on the organisation and evolution of the MHC as demonstrated here by the example of a 200 kb contig that has been analysed for local and global features. In conjunction with cross-species comparisons, our results present new evidence on the structure of isochores, the evolutionary dynamics of repeat-mediated recombination and its effect on certain MHC encoded genes, and a higher than average degree of natural polymorphism that has implications for sequencing the human genome. We also report the finding of a class I-related pseudogene (HLA-ZI) in the middle of the class II region, which provides the first direct evidence for DNA exchange between these two related regions in man.

DNA sequencing of the MHC class II region and the chromosome 6 sequencing effort at the Sanger Centre.

The human Major Histocompatibility Complex (MHC) is located on the short arm of chromosome 6 (6p21.3) and spans about 4 Mb. According to different gene families the MHC is subdivided into a class I, class II and class III region and many of its gene products are associated with the immune system and the susceptibility to various diseases. To date, we have sequenced about 40% (400 kb) of the class II region between HLA-DP and HLA-DQ and a coordinated effort to sequence the entire MHC is well underway. Analysis of the sequence revealed several novel genes and provides new insights into the molecular organisation and evolution of the MHC. All our data are publicly available via the MHC database (MHCDB) which allows rapid access, retrieval and display in the context of other MHC associated data. MHCDB is online available at (http:(/)/www.hgmp.mrc.ac.uk/) and, together with all our sequences also via anonymous ftp (ftp.icnet.uk/icrf-public).

Organisation and functions of class II genes and molecules.

The class II region of the human MHC contains all of the known class II genes: as well as antigen processing components and only one gene not obviously associated with the immune system, RING3. As an approach to understanding linkage disequilibrium and recombination in relation to polymorphism of the region we are cloning and sequencing the class II region. To date, the sequence of the DP-DQ region has almost been completed (see Report by S. Beck). Several sets of genes implicated in the immune system, especially in antigen processing and presentation, are clustered together in the MHC: class I (HLA-A, B, C etc) class II (DR, DQ, DP, DN, DO, DM) LMP2 and 7, TAP1 and 2, TNF, C2, C4, Bf, Hsp70. This situation has provoked speculation that the MHC behaves as a gene cluster in which allelic products of polymorphic genes are maintained on a haplotype so as to co-ordinate T cell repertoire development and deployment. The high levels of linkage disequilibrium across the region are consistent with this idea. Functions of the genes in the MHC are being investigated as a step towards gaining insight into antigen processing and presentation as well as understanding MHC-disease associations. We are concentrating on the functions of the class II-related genes, DM and DN/DO as well as the TAP/LMP cluster.

Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease.

Some cases of Alzheimer's disease are inherited as an autosomal dominant trait. Genetic linkage studies have mapped a locus (AD3) associated with susceptibility to a very aggressive form of Alzheimer's disease to chromosome 14q24.3. We have defined a minimal cosegregating region containing the AD3 gene, and isolated at least 19 different transcripts encoded within this region. One of these transcripts (S182) corresponds to a novel gene whose product is predicted to contain multiple transmembrane domains and resembles an integral membrane protein. Five different missense mutations have been found that cosegregate with early-onset familial Alzheimer's disease. Because these changes occurred in conserved domains of this gene, and are not present in normal controls, they are likely to be causative of AD3.

Released chromatin: linearized DNA for high resolution fluorescence in situ hybridization.

Free DNA was prepared from routinely harvested and fixed cells for high resolution FISH mapping using either a sodium hydroxide/ethanol mixture or 70% formamide. Hybridization signals from cosmid probes appeared as extended lines. The average length of signals on DNA prepared with sodium hydroxide was significantly greater than with formamide. A set of overlapping cosmids from the HLA class II region was used to determine how precisely the actual overlap or gap between probes can be calculated from the measured overlap or gap between their signals. Lengths of the probe signals and their known kilobase lengths were used as an internal ruler. The mean values calculated from the measured length from 30 or more signals for each probe pair showed remarkable conformity with the known kilobase lengths of their overlaps and gaps. Immediately adjacent probes could also be ordered on the released DNA. These simple procedures dramatically increase the speed with which relationships between probes can be determined during contig construction.

Cloning and characterization of human phosphatase inhibitor-2 (IPP-2) sequences.

cDNA clones similar to rabbit muscle phosphatase inhibitor-2 (IPP-2) were isolated from human libraries. On Northern blots two transcripts of approximately 2kbp and approximately 4kbp were detected in all tissues tested. Analysis of cDNA sequences showed that the longer transcripts were similar to the shorter clones but contained extended 3' ends. The human nucleotide sequence was highly homologous (94% identity) to the rabbit IPP-2 sequence and encoded a peptide of 205 amino acids. IPP-2 sequences were highly conserved throughout vertebrates. Southern hybridization results were consistent with the existence of a family of related IPP-2 sequences in the human genome. Most of these are likely to be pseudogenes, since all of the cDNA clones examined could have originated from a single gene. By in situ hybridization IPP-2 sequences were mapped to several different human chromosomes. We sequenced one gene located in the major histocompatibility complex (MHC) on Chromosome (Chr) 6 that contained the entire coding region of IPP-2.

Cloning of the region between HLA-DMB and LMP2 in the human major histocompatibility complex.

The human MHC is one of the most extensively mapped regions of the human genome. Almost all of the class II region of the MHC has already been cloned in cosmids but a gap remained between the DMB and LMP2 genes. Previously, screening of several complete cosmid libraries had failed to bridge this gap, which may contain novel antigen processing or presentation genes. We constructed cosmid libraries from two different sources in order to clone the region: (a) a library with fourfold coverage made from flow-sorted human chromosome 6 DNA and (b) a library derived from a yeast artificial chromosome clone spanning the region. Using this saturation approach, cosmid clones were eventually isolated over the region of interest. A single bacteriophage P1 clone was also obtained spanning the region. The YAC, cosmid, and P1 physical maps were consistent and the distance between the DMB and LMP2 genes was measured as 70 kb. It is not clear why DMB to LMP2 is infrequently represented in cosmid libraries, but the clones that we have obtained will now enable us to search for new coding sequences.

Two simple procedures for releasing chromatin from routinely fixed cells for fluorescence in situ hybridization.

We describe two methods for releasing chromatin from routinely harvested and fixed cells. Using fluorescence in situ hybridization with combinations of probes from the HLA class II region, we show that good signals can be obtained on free chromatin fibers enabling determination of relationships between closely adjacent or overlapping probes.

Efficiency and specificity of gene isolation by exon amplification.

Exon amplification is an increasingly popular approach to the identification of transcribed sequences and will complement other strategies to isolate genes. We have used this system to amplify candidate exons from 32 cosmids, including 8 cosmids which span a well characterized 185-kb region of the human major histocompatibility class II region on Chromosome (Chr) 6. We have examined the efficiency, specificity, and reproducibility of the system in isolating exons from genes known to be present on particular cosmids and have determined the nature and frequency of artefact amplifications in routine cosmid screening. We were able to clone at least one exon from 88% (7/8) of all known genes tested (including exons which are differentially spliced) and obtained artefacts from 19% (6/32) of the cosmids tested. Such artefacts generally arise from the amplification of noncoding sequences flanked by regions with high homology to acceptor and donor splice junctions. We show that the exon amplification procedure can be used successfully with a wide variety of cosmids which have different numbers of genes and gene structures and describe several approaches to the characterization of novel exons cloned in this study.