ESHG Posters 11

A. Aleyasin¹, W. Barendse²;
¹National Research Center for Genetic Engineering and Biotechnology, Tehran, IRAN (ISLAMIC REPUBLIC OF), ²TAG, CSIRO, Brisbane, Qld, AUSTRALIA.

Human chromosome 22 is one of the smallest autosomes involved in many human genetic disease. Comparative mapping of human chromosome 22 has not been very well described in most unrelated species. This work was to choose homologous genes from human chromosome 22 (HSA22) and map them in cattle to understand the homology between human chromosome 22 and cattle chromosomes. In this way polymorphisms for six genes (ADSL, BZRP, CYP2D, CRYBA4, DIA1 & PDGFB) that showed strong sequence homology with their human counterparts and their locations are well known in human have been developed in cattle. Polymorphisms were developed for introns or 3'UTRs of those genes. Those polymorphisms have been genotyped and genetically mapped in cattle using International Bovine Reference panel containing individuals related to three generations. Those genes have been mapped by genetic linkage analysis in the frame work provided by International cattle genome mapping facility in CSIRO, Brisbane Australia. They have been mapped on cattle chromosome 5 and 17 and developed understandings of conservation between human chromosome 22 and cattle chromosome 5 and 17.

Molecular characterisation of pericentric inversion breakpoints on chimpanzee chromosome 19 compared to human

H. Kehrer-Sawatzki¹, S. Tänzer², M. Platzer², H. Hameister¹;
¹Department of Human Genetics, University of Ulm, GERMANY, ²Department of Genome Analysis, Institute for Molecular Biotechnology, Jena, GERMANY.

Structural differences of the primate chromosomes include heterochromatin variability, inversions, and the telomeric fusion of two ancestral chromosomes, which resulted in the generation of human chromosome 2. Since genetic disparities are assumed to be responsible for phenotypic differences between hominoid species, molecular characterisation of the genomic regions harbouring evolutionary breakpoints is important. These analyses will help to understand the mechanisms involved in chromosome evolution. Additionally, the molecular definition of evolutionary breakpoints will reveal whether they overlap with human genomic regions susceptible to constitutional or somatic rearrangements and whether owing to intrinsic sequence features these breakpoints have been utilised more than once during evolution. So far, only the fusion event, which gave rise to human chromosome 2 has been characterized on the molecular level. Currently, we are investigating the pericentric inversion of the chimpanzee (PTR) chromosome 19, orthologous to human chromosome 17. Human BAC clones from 17p13 and 17q21.3 were identified, which span the inversion breakpoints. FISH with these BACs resulted in split hybridization signals on PTR19. Subfragments of these human BACs were used to isolate chimpanzee BACs, which in turn cover the inversion breakpoints. Sequence analysis of the junction and the respective normal sequences revealed that the inversion took place in 5-7kb spanning regions rich of repetitive elements and that nonhomologous recombination between Alu elements has been involved. We specified the sequence environment flanking the breakpoints, identified the genes next to the breakpoints on either site, and addressed the contribution of this inversion to the structure of these genes.

Y. Karavidopoulou, S. Contrino, P. Kersey, M. Magrane, A. Williams, R. Apweiler;
European Bioinformatics Institute (EBI), Cambridge, UNITED KINGDOM.

TrEMBL was created as a supplement to the SWISS-PROT Protein Knowledgebase. Whereas SWISS-PROT is a non-redundant, manually curated protein sequence database, TrEMBL is a computer-annotated database containing the translations of all coding sequences present in the EMBL-Bank nucleotide sequence database that are not present in SWISS-PROT.
SWISS-PROT contains 7,652 annotated human sequences. 31,673 exist in TrEMBL awaiting annotation and/or merging. We aim to fully annotate and describe these human protein sequences and distribute them to the scientific community as soon as possible, thus enriching the biological knowledge of the human genome. More about the 'Human Proteomics Initiative' (HPI) at http://www.ebi.ac.uk/swissprot/hpi/ .
A human SWISS-PROT/TrEMBL non-redundant proteome set was built, consisting of 24,049 entries (from a total of 39,325 human entries). Proteome analysis was performed, offering statistical, structural, functional and comparative analyses, http://www.ebi.ac.uk/proteome/ .
As all human coding sequences are not yet in the nucleotide sequence database, the SWISS-PROT and Ensembl teams jointly constructed a draft complete non-redundant proteome, comprising 23,377 SWISS-PROT/TrEMBL non-redundant entries and additional 9,136 Ensembl predictions.
Gene sets for the human chromosomes were constructed, providing a comprehensive reference of the human data in SWISS-PROT/TrEMBL. Each set is an alphabetic listing of the genes with the HUGO approved gene symbol (or the NCBI LocusLink provisional symbol) encoded on that chromosome, together with the chromosome position, the protein it encodes and useful links to other databases, such as GeneCards, OMIM, Ensembl, InterPro and CluSTr.
Currently, 14,932 gene mappings have been made, 8,427 of which with an official HUGO gene name.

Biomax PEDANTÔ Human Genome Database - Automatic and Manual Functional Annotation of the Human Genome

A. Fritz¹, B. Geier¹, V. Solovyev², M. Firsov³, C. M. E. Schüller¹;
¹Biomax Informatics AG, Martinsried, GERMANY, ²Softberry Inc., Mount Kisco, NY, ³Petrogen Ltd., St. Petersburg, RUSSIAN FEDERATION.

Public and private efforts have begun to extract valuable information from the wealth of human genome data. Biomax Informatics systematically analyzed data for all human genes (both known and unknown) in a consistent manner to determine (possible) protein function and, when necessary, refined the annotation manually . All information has been stored in a relational database, which can be accessed via a Web-based user interface.
Using the publicly available working draft assembly of the human genome, we identified the location of known and putative genes with the exclusive Fgenesh++ software from Softberry, Inc. The Pedant-ProÔ Sequence Analysis Suite from Biomax was used to perform systematic, comprehensive and consistent analysis for in-depth functional and structural characterization of the predicted proteome. This characterization includes functional class assignments according to a functional catalogue (with more than 1500 functional categories) and assignment of EC numbers, PROSITE patterns, Pfam domains and SCOP classifications.
The PEDANTÔ Human Genome Database based on the 12 December 2000 working draft assembly of the human genome and contains 44,403 gene models. Using sequence similarity to manually annotated proteomes, proteins have been assigned to functional categories. In addition, the Pedant-Pro software performed further functional and structural analyses including similarity searches, annotation of functional domains, and assignment of keywords, EC numbers and transmembrane domains. Report pages display all important automatic and manual annotations for each protein and hyperlinks to the corresponding databases. The easy-to-use graphical user interface provides search tools as well as DNA and protein viewers.

The dynamic process of gene discovery: characterization of 19 novel transcripts from human chromosome 21

S. Deutsch¹, A. Camargo², A. Reymond¹, B. J. Stevenson³, R. B. Parmigiani², C. Ucla¹, F. Bettoni², C. Iseli³, V. Jongeneel⁴, P. Bucher³, A. J. G. Simpson², S. E. Antonarakis¹;
¹Division of Medical Genetics, University of Geneva Medical School, Geneva, SWITZERLAND, ²Ludwig Institute for Cancer Research, Sao Paulo, BRAZIL, ³Swiss Institute of Bioinformatics, Epalinges, SWITZERLAND, ⁴Ludwig Institute for Cancer Research, Epalinges, SWITZERLAND.

The identification of all human chromosome 21 (HC21) genes is a necessary step in understanding the molecular pathogenesis of trisomy 21 (Down syndrome). The first analysis of the sequence of 21q included 127 previously characterized genes, and predicted an additional 98 novel anonymous genes. Recently we evaluated the quality of this annotation by characterizing a set of C21orfs and PREDs, identified by mapping spliced ESTs to the genome and only by in silico analysis, respectively. This study underscored the limit of in silico-only gene prediction, as many PREDs were incorrectly predicted. To refine the HC21 annotation, we have developed a reliable algorithm to extract and stringently map sequences that contain bona fide 3’ transcript ends (3’-tags) to the genome. We then created a specific 21q Acedb that incorporates new ESTs as well as features such as CpG islands, repeats and gene predictions. Using these tools we identified 27 new putative genes. To validate these, we sequenced previously cloned cDNAs, performed RT-PCR, 5’ and 3’RACE procedures and comparative mapping. These approaches substantiated 19 new transcripts, thus increasing the HC21 gene count by 9.5%. These transcripts were not previously identified probably because they are small and encode small proteins. We also identified 4 transcriptional units that are spliced but contain no obvious open reading frame. The HC21 data presented here further emphasize that current gene predictions algorithms miss a substantial number of transcripts that nevertheless can be identified using a combination of experimental approaches and multiple refined algorithms.

C. Mugnier¹, M. Chauvet², K. Tanin², F. Barbet², A. Munich³, M. Roux-Rouquie⁴, J. Frezal⁵;
¹Université de Paris V, Paris, FRANCE, ²Université de Paris V - Génopole Ile de France, Paris, FRANCE, ³U393 INSERM - Université de PARIS V - Génopole Ile de France, Paris, FRANCE, ⁴Institut PASTEUR, Paris, FRANCE, ⁵Hôpital Necker - Université de Paris V - Génopole Ile de France, Paris, FRANCE.

We designed a new conceptual model to improve data storage and retrieval in GENATLAS
Data structure in GENATLAS_TM : The new relational GENATLAS, noted GENATLAS_TM, was developed with ORACLE and contains five major sections which consist in the GENATLAS_TM core. The section GENE stores information on DNA type, arrangement and structure for a specific gene or DNA sequence. The section RNA contains data on the major gene transcript type and isoforms as well as tissue expression or associated pathologies (stored in the special section VARIANT/PATHOLOGY). Protein products are categorized according to their structures: secondary structure, structure motifs/domains, structure homology, etc., and their activity categories in the PROTEIN section. In addition, the protein EXPRESSION section displays ontologies about cellular localization (extra- and subcellular), tissue type, stage, etc.
The GENATLAS_TM core is linked to three distinct directories which are the CITATION directory (containing more than 40.000 occurrences linked to Medline abstracts), the PHENOTYPE directory gathering information of mendelian disorders, susceptibility genes, somatic genetic disorders (tumors, malformations, etc.) and the LINKAGE directory includes 28.000 pairs At the moment, GENATLAS_TM contains more than 12.000 genes, and 2.600 diseases are instantiated.
Querying GENATLAS_TMThe strategies to query GENATLAS_TM implement multicriterion approaches according to the core sections (GENE, VARIANT/PATHOLOGY, RNA, PROTEIN, EXPRESSION) or particular fields within these sections. Actually, criterion combinatory querying constitutes one of main features of GENATLAS_TM; more than one hundred criterions can be used to query for gene structure, function and diseases.
GENATLAS_TMwould be accessible at : http://www.dsi.univ-paris5.fr/genatlas

In man trinucleotide repeats are underrepresented but some motifs are overexpanded: a novel signature for eukaryotic genomes?

P. Astolfi¹, D. Bellizzi², V. Sgaramella²^,3;
¹Dept. Genetics and Microbiology, University of Pavia, Pavia, ITALY, ²Dept. Cell Biology, University of Calabria, Cosenza, ITALY, ³Centro Linceo Interdisciplinare, Rome, ITALY.

Aiming at assessing whether the reported shortage in trinucleotide repeats could be ascribed to specific motifs, such as those found dynamically mutable in man, we used the Tandem Repeat Finder program and detected approximate di-, tri- and tetra-repeats in human chromosomes 21 and 22 and in five organisms (M. musculus, D. melanogaster, C. elegans, A. thaliana, S. cerevisiae). Di-repeats are always the most represented, tetra-repeats are more represented in mammals than in the other organisms, whereas tri-repeats are consistently scarce. In man tri-repeat representation is less than ¼ of tetra-; their frequency is 6.4 repeat/Mb and their coverage 473 b/Mb. Motifs such as ACG, ACT, CCG are scarce, while the others present taxa-specific variations. Motif representation seems related to base sequence rather than content, with AA contributing positively and CG negatively. In man ACG is the least represented, ATC has the highest coverage, AAT the most frequent. Tri-repeat coverage generally increases linearly with frequency, except for AAG, ACC, AGG, ATC in man, and AAG, AGG in mouse: these repeats tend to expand in the former more than in the latter and occasionally, but never in exons, exceed copy number of 40, the observed limit for tri-repeat pathological expansions. Tri-repeat scarcity seems a feature of mammals and the variable representation of their motifs may constitute a novel signature for eukaryotic genomes: since both probably result from a structural control on DNA synthesis, the aberrant elongations found in dynamic mutations may be due to the derangement of one or more control elements.

DNA-Protein Interaction DataBaseProject: Bioinformatic DataBase of Third Genertion.

S. A. Vasil'ev¹, A. V. Alexeevski¹, S. A. Spirin¹, V. N. Tashlitski², A. S. Karyagina-Zhulina³, Y. Kalaidzidis¹;
¹Belozersky Institute of Physico-Chemical Biology, MSU, Moscow, RUSSIAN FEDERATION, ²Chemical department, MSU, Moscow, RUSSIAN FEDERATION, ³Institute of Agricultural Biotechnology RAAS, Moscow, RUSSIAN FEDERATION.

Increasing significance of theoretical modeling in the field of molecular medicine and molecular biology makes databases (DB) for three-dimensional (3D) structures of biological macromolecules particularly important. Currently available DB for 3D structures of biological macromolecules can be divided by three generations. First generation (G1) is DB-depositories. To the moment Protein Data Bank (PDB) is the most successful and actively developing G1 DB. Second generation (G2) is specially designed DB. Nucleic Acid Database (NDB) is an example of G2 bioinformatics DB (BDB). Along with information stored in PDB this type of DB allows linking to other biological DB (PFAM; GenBank, etc.). Third generation (G3) is highly specialized informational systems (IS). This type is presented by classified IS like Protein-DNA Interaction and PDBSum projects that provide both detailed classification and analysis of the information stored in PDB and NDB. Currently we develop G3 BDB -DNA-Protein Interaction DataBase (DPIDB) version 3.0.1. based on Pluk (object oriented) technology.
DPIDB contains complete information on structure of DNA-protein complexes (derived from PDB and NDB) in both usual and specially designed by us data format. Our data format allowed revealing specific features of DNA-protein interaction in complexes. The up to date classification of DNA-protein complexes is used for DPIDB. The Internet site of DPIDB v 3.0.1 (http://www.dpidb.belozersky.msu.ru) is now under construction. It will include DPIDB as well as the results of our own researches, for example, the statistical analysis of DNA-protein interactions. To the moment a review regarding two DNA-binding protein groups are presented on the site.

Aggregation of post-transcriptionally coregulated mRNAs in cytoplasmic ribonucleoprotein particles

A. Totaro, M. D'Apolito, A. Notarangelo, L. Zelante, A. Quattrone;
IRCCS CSS, S. Giovanni Rotondo, ITALY.

A cytoplasmic ribonucleoprotein (mRNP) infrastructure works in the cell as an interface between the transcriptome and the proteome, being responsible for transcript transport, localization, turnover and translational control. This complex is composed by aggregates of distinct subsets of mRNAs with regulatory RNA binding proteins, which finally interact with ribosomes and the cytoskeletal apparatus to allow protein synthesis. It has been proposed that the clusters of monocistronic mRNAs composing individual mRNPs function like poycistronic mRNAs transcribed from operons in procariotic organisms, adding with their combinatorial nature a further level of regulative complexity. If these mRNPs are really regulative units, the bound mRNAs should be endowed with uniform turnover and translation rate. Here we show that a clusterization of human primary fibroblast mRNAs based on their individual half life (measured by an high density microarray approach) produces groups of mRNAs which can be demonstrated to be colocalized, and probably belong to the same mRNP type. The understanding of this new level of cell organization has profound implications for the pathogenesis of those mendelian diseases, like myotonic dystrophy and X fragile syndrome, in which a single alteration in a mRNP component results in a complex clinical phenotype.

T. M. Strom¹^,2, C. Glöckner¹^,2, T. Meitinger¹^,2;
¹Institute of Human Genetics, GSF National Research Center, Munich, GERMANY, ²Institute of Human Genetics, Klinikum rechts der Isar, Technical University, Munich, GERMANY.

We constructed a database using dbSNP and the GoldenPath data which can be accessed via a Web-browser (http://ihg.gsf.de). The search tool allows to retrieve SNPs in the genomic region of RefSeq and Ensembl transcripts and to download the genomic sequence around each SNP as given by the GoldenPath. SNPs were masked by an 'N' in the genomic sequence in order to avoid primers to be positioned across SNPs.
The database contains all entries of dbSNP mapped to the GoldenPath by the NCBI. Entries are only displayed when they hit a single chromosome, a single contig and maximally twice within a certain contig. The proportion of repeat sequence within the region -100 bp to +100 bp was calculated for each SNP. The number of exons for each gene is derived from the GoldenPath annotation.
The database can be searched by i) chromosome coordinates, ii) RefSeq or Ensembl accession numbers, or iii) rsSNP IDs. A non-redundant dataset of RefSeq and Ensembl transcripts is used for the chromosome search. All SNPs up to 10,000 bp upstream and downstream of the transcripts are displayed. Transcripts without SNPs are indicated. The search can be narrowed by the proportion of repeats and the average estimated heterozygosity.
The tool allows seven features to be downloaded: i) rsSNP ID, ii) RefSeq or Ensembl accession number, iii) chromosome position on the GoldenPath, iv) average estimated heterozygosity, v) proportion of repeats, vi) observed polymorphism (dbSNP) and vii) sequence around the SNPs either as given in dbSNP or in the GoldenPath.

C. Beroud¹, R. Ben Yaou², A. Helbling-Leclerc², S. Llense¹, J. C. Kaplan¹, F. Leturcq¹, D. Recan¹, G. Bonne²;
¹Institut Cochin de Génétique Moléculaire, Paris, FRANCE, ²INSERM U523 - Institut de Myologie, Paris, FRANCE.

About 75 neuro-muscular disease genes have been identified so far. Among them, Emery-Dreifuss syndrome is due to defects in the emerin gene (STA on chromosome X), or in the lamin A/C gene (LMNA on chromosome 1), both coding for proteins associated to the nuclear envelope. A large clinical intra- and inter-family variability is associated to both emerinopathy and laminopathy. In addition, up to five different clinical entities are associated to LMNA mutations. To handle and exploit the numerous data collected by the “French Clinical and Research network for Emery-Dreifuss muscular dystrophies and other nucleopathies" connected to "EUROMEN" network on emerinopathy and laminopathy (a subsidiary of the “Myo-Cluster” EC initiative), we developed a Locus Specific DataBase (LSDB). It was adapted from the Universal Mutation Database (UMD) package previously used for many LSDB's such as p53, APC, FBN1, LDLR, VHL, MEN1, ATP7B... The specific features of the present software are: (i) access to the various routines via internet; (ii) inclusion of multi-parametric analytic tools allowing optimized searching of correlations. This requires a full implementation of relevant clinical, para-clinical and biological features. We have developed a specific module to display phenotype-genotype and genotype-phenotype correlations. To date the UMD-STA database contains 130 records and the UMD-LMNA database 250 records.
This work was supported by the AFM/INSERM French Network "Clinical and research network for Emery–Dreifuss muscular dystrophies and other nucleopathies" and the European "Myo-Cluster/EUROMEN" Network.

HC Forum® is the first secure and generic informatic platform dedicated to medical genetics, in which information is structured according to pedigrees and the different components of medical genetics. It includes a database accessible from a Web site. User-friendly interfaces require no specialised computer skills, but only an access with a simple Web navigator. Graphic tools allow inputting and search for information in the database. A high level of security includes authentication with smart cards, certified high level encryption (128 bits), medical data coding to avoid any patronymic names, and digital tattooing of images.
HC Forum® is approved by the French "Informatics and Freedom Commission " and by security audits, a medical ethics laboratory (Pr C. HERVE, Paris) and by specialized lawyer.
HC Forum® offers geneticists private working spaces dedicated to cytogenetics, monogenic diseases and dysmorphological syndromes, including LDDB thesaurus (R. WINTER) and international nomenclatures. It allows the inputting of genetic records with pedigree and a standardised textual description of diseases, enriched with images from the different clinical and paraclinical investigations.
Furthermore, HC Forum® allows collaborative working: (i) shared records allow several doctors to work on the same record, (ii) thematic networks allow professionals to share data related to a same disease.
HC Forum® establishes a link between medical genetics departments, research laboratories and teaching centres. HC Forum® is academic, free of charge and available to the international scientific and medical communities.
URL: https://HCForum.imag.fr

Genomic Assays Project: Building an SNP-Based Linkage Disequilibrium Map and Ready-to-Use 5’ Nuclease Assays for 200,000 SNPs

P. Dong, F. de la Vega, E. Spier, J. Stevens, K. Guegler, D. Dailey, G. Nunn, L. Wogan, J. Ziegle, M. Rhodes, D. Madden, D. Gilbert;
Applied Biosystems, Foster City, CA.

The human genome project has identified the 3 billion base human genome sequence in the past decade. SNPs in the human genome, encoding the human variance, are expected to become a powerful tool in deciphering complex traits such as disease susceptibilities and drug responses.
We describe the project to create a ready-to-use solution for association studies in the process of gene discovery. We are building a genome wide linkage disequilbrium (LD) map based on data from both the Celera and the public sequencing efforts. The SNPs selected for the LD map are validated in regards to both assay performance and allelic frequencies in at least two populations. During the project, we will make available 200,000 SNP ready-to-use assays named Assays-On-DemandTM.
All SNP assays created in this project will be of high information content, quality and performance, based on the process we have introduced. These SNP assays will be easily accessible via a simple interface, enabling researchers to assemble the list of SNP assays tailored for their individual projects.
We will be presenting details of the genomic assays project including SNP selection , assay validation and LD map creation. The ultimate goal of this project is make genetic studies faster and more efficient.

Comparative functional genomics of an entire chromosome: coding and regulatory annotation of HC21 using the mouse genome

E. T. Dermitzakis¹, A. Reymond¹, S. Deutch¹, B. J. Stevenson², C. Iseli², P. Bucher², V. Jongeneel², S. E. Antonarakis¹;
¹Division of Medical Genetics, University of Geneva, Geneva, SWITZERLAND, ²Ludwig Institute for Cancer Research, Lausanne, SWITZERLAND.

Chromosome 21 (HC21) is the smallest human chromosome. Several diseases are associated with genes or genomic regions of HC21 including Down syndrome, the most common cause of mental retardation. The availability of the mouse genomic sequence gives the opportunity to combine whole chromosome homologous sequence comparisons with experimental assays to reveal novel functional regions that correspond either to putative genes or regulatory regions. We used PipMaker to perform long genomic alignments of the entire human chromosome 21q sequence with the homologous sequences from mouse chromosomes 16, 17 and 10 obtained from the Celera mouse genome assembly. 3896 blocks of >100bps in length and >70% sequence identity were extracted and fragments corresponding to known HC21 exons were excluded. From the remaining 2383 unassigned conserved blocks, 10% correspond to deposited usually singly represented human ESTs, 37% contain significant ORFs and several have strong similarity to deposited coding sequences. These suggest that many of the conserved blocks are unknown exons of known or unknown genes that were neither experimentally identified nor predicted by the sequence analysis of HC21, probably due to short length and restricted spatial and temporal expression. Conserved blocks corresponding to putative exons or regulatory regions are currently being tested experimentally for their functional role. Additionally, coding and intergenic sequences are being analyzed in order to reveal their evolutionary properties. We will present a combination of comprehensive computational and experimental analysis in an effort to improve and expand the coding and regulatory annotation of HC21 and study its evolutionary history.

Transcriptome variations in Caco-2 cells during their differentiation : a model for studying intestinal iron absorption ?

H. Ferran Bédrine¹, I. Gicquel¹, M. Lecunff², I. Guisle², N. Soriano¹, P. Fergelot¹, J. Y. Le Gall¹, J. Leger², J. Mosser¹;
¹UMR 6061CNRS, Rennes, FRANCE, ²INSERM U533, Nantes, FRANCE.

Iron homeostasis is controlled by intestinal absorption. Many factors implicated in this absorption have been recently characterized without explaining its adaptation to the iron store of the whole organism. This adapted absorption is probably due to an unknown sensibilization of intestinal crypts undifferentiated cells. These cells differentiate during their migration along the crypt-villosity axis and would become fully differentiated enterocytes with an appropriate iron absorption capacity. Caco-2 cells, a colon adenocarcinoma cell line, seem to be a good model for studying iron effects on gene expression during the enterocytic differentiation.
A transcriptomic approach has been implemented to identify genes involved in the regulation of iron absorption in the intestine : 1536 cDNA were selected by subtractive suppressive hybridization ( SSH ) and spotted on micro arrays. Their sequences reveal that each array represents 700 genes, half of them with an unknown function. mRNA from Caco-2 cells at different stages of differentiation were hybridised to the microarrays. These experiments show that 400 out of the 700 genes can be classified in seven distinct expression profiles that seem to correlate with functional role, most of iron metabolism genes being indeed overexpressed in differentiated cells. The effects of intracellular iron concentration on the expression of those genes are currently being studied. These observations should enable us to better understand the mechanisms underlying primary iron overload. (genetic heterogeneity, incomplete penetrance of the C282Y mutation of the HFE1 gene.)

The use of multiple algorithms and a novel quality metric system for high throughput and accurate microsatellite- and SNP-based genotyping

D. C. Siu, D. Bishop, T. Hatch, C. Frantz, Y. Lou, M. Rhodes, C. Brown;
Applied Biosystems, Foster City, CA.

With the advent of automated fluorescent genotyping using medium-to-high throughput systems such as the ABI PRISM ® 3700 and 3100 Genetic Analyzers, a fast and accurate genotyping software tool is needed to perform accurate allele calling with minimal human interventions.
The ABI PRISM® GeneMapperÔ software, containing a multiple algorithm module and a novel quality metric system, was developed to meet the requirements of these automated, microsatellite- and SNP-based genotyping processes. In GeneMapperÔ software, an algorithm module containing multiple allele calling algorithms - size calling, binning, allele calling, has been developed to optimize the accuracy of final genotype assignments. Coupled with a novel Process Based Quality Value (PQV) system, which contains over 20 quality metrics to pinpoint the origins of genotyping failures, GeneMapper software enables users to choose to manually examine only allele calls with low quality scores. For projects that require frequent modifications of markers, a novel panel management feature has been developed to allow on-the-fly editing of bins within markers.
A collection of genotyping data has been processed using GeneScan® software, GenotyperÔ software and GeneMapperÔ software, and analysis results were compared. Allele calling accuracy and time required for complete analysis and data editing will be presented.

M. D. Rhodes, S. Eddins, A. Dodge, K. Rogers, D. Dailey, P. Dong, D. Madden, D. Gilbert;
Applied Biosystems, Foster City, CA.

The draft of the human genome enables scientific discoveries in many fields including gene expression and genotyping. The challenge is to make this wealth of genome information available to the laboratory. Applied Biosystems has launched two complementary product offerings that together provide a complete solution for ready to use assays for quantification of gene expression and genotyping via SNPs These assays use the 5‘ nuclease assay with Taqman® MGB probes. All assays are provided in an easy to use single tube format. Experimental setup is simple consisting of addition of two reagents (assay mix and universal master mix) to the Target DNA.
The Assays-by-Design^SM service, will allow customers to submit a target sequence of interest. Applied Biosystems uses proprietary algorithms to design an assay for each target. Following synthesis and formulation, the assay will be analytically tested for integrity before shipping to the customer. The second offering, the Assays-on-Demand^TM products, will provide ready-to-use assays to pre-defined targets which customers may select by, SNP or transcript ID, gene name or attribute. Genomic assays provided by both Both Assays-by-Design^SM and Assays-on-Demand^TM are provided in an easy to use single tube format. Experimental setup is minimized consisting of addition of two reagents (assay mix and universal master mix) to the Target DNA. The process is highly automatable and signal detection can be carried out unattended using the ABI Prism® 7900HT with robotic accessories
Details on the design, production and testing of these assays will be presented.