Sponsored by the U.S. Department of Energy Human Genome Program
Human Genome News Archive Edition
Human Genome News, March 1994; 5(6)
The Third International Workshop on the Identification of Transcribed Sequences, held in New Orleans October 2-4, 1993, was sponsored by DOE, NIH, and Amgen, Inc. Some 60 scientists from 9 countries discussed new approaches for identifying genes in genomic material, experiences and challenges in applying these methods, and preliminary data on assembling transcriptional maps.
The two major ways of assembling transcriptional maps of the human genome are to (1) isolate and sequence all expressed mRNAs as cDNA clones and place them on the physical and genetic maps or (2) identify all transcribed sequences in cloned mapped genomic DNA. Three different approaches can be used for the latter method: interpret genomic sequence with the help of computer programs, assay genomic sequences functionally for their ability to splice out exons in exon-trap protocols, and hybridize genomic DNA to cDNA. Although all hybridization methods are conceptually based on the same process, a number of technical variations exist. For example, the genomic partner can be yeast artificial chromosome (YAC) clones, cosmids, or parts of chromosomes; both partners in the hybridization reaction can be in solution or one can be fixed on a solid support; and the cDNA can be eluted and cloned or used to score signals on genomic clones.
Three new variations on hybrid selection were presented: a coincidence-cloning protocol involving a selective ligation step (Anthony Brookes, Medical Research Council Human Genetics Unit); a hybrid-selection protocol based on the polymerase chain reaction (PCR) and involving a genomic-cDNA chimera (Pudur Jagadeeswaran, University of Texas); and a selection through genomic RNA and single-stranded cDNA (Anand Swaroop, University of Michigan). John Hozier (Applied Genetics Laboratories, Inc.) described pilot experiments on hybrid selection using metaphase chromosome spreads as the genomic partner and cloning cDNAs eluted from microdissected parts of chromosomes. Bento Soares (Columbia University) combined the hybrid-selection method with the direct cDNA-screening approach by hybridizing cDNAs to filters carrying genomic clones from entire chromosomes, making cDNA-genomic hybrids visible through a sandwich technique, and eluting the cDNAs from positive genomic clones.
A new version of the pSPL-1 exon-trap vector, pSPL-3, was described by Paul Nisson (Gibco BRL/Life Technologies, Inc.). Nicole Datson (Leiden University) reported progress in using the pETV-SD2 exon-trapping vector.
Advances in three neural network approaches for gene identification in genomic sequences were presented for GeneParser (Eric Snyder, University of Colorado), GeneID (Roderic Guigo, Los Alamos National Laboratory), and GRAIL (Richard Mural, Oak Ridge National Laboratory). GRAIL now recognizes smaller exons, and the Gene Assembly Program assembles full-length coding sequence from genomic sequence with high reliability.
Techniques to identify and sequence cDNAs were presented. James Eberwine (University of Pennsylvania), Mark Erlander (Scripps Research Institute), and Wai-Choi Leung (Tulane University) discussed identifying cDNAs by differential display techniques in single cells, different areas of the brain, and different tissues, respectively. Radoje Drmanac (Argonne National Laboratory) and Joachim Rothe [Imperial Cancer Research Fund (ICRF)] reported on sequencing by hybridization. James Sikela (University of Colorado) addressed sequencing full-length cDNA clones by using NcoI sites in 5' ends of cDNA clones.
A number of groups presented different versions of the hybrid-selection approach. Although use of whole-cell DNA from YAC-containing yeast as the genomic partner is possible (Sherman Weissman, Yale University), other investigators have used purified YACs or cosmids as the genomic partner. After exclusion of ribosomal and repeat-sequence contaminants, 60 to 90% of cDNAs map to the genomic region.
Improvements were reported in direct cDNA screening of arrayed genomic libraries. Sensitivity was increased by using a subtracted cDNA probe (Jeffrey Falk, Scripps Research Institute), and GRAIL allowed investigators to proceed quickly from identified genomic clones to coding sequences (Wolfgang Schwabe, National Institute of Mental Health).
Results on exon trapping with the pSPL vector system were reported for three different genomic regions (Ken Abel, University of Michigan; Michael North, ICRF; Marie-Laure Yaspo, ICRF). Although exon trapping on YACs is possible, individual or pools of cosmids seem to be the preferred unit for exon trapping because of skewing in the PCR step. A difficulty is verifying that a trapped product is an exon. Their generally small size frequently makes trapped products ineffective probes for zoo blots or cDNA library screenings.
In addition to these newer methods for gene identification, successes were reported with more-standard procedures: cloning CpG islands (Daniela Toniolo, Consiglio Nazionale delle Ricerche), identifying evolutionarily conserved sequences (Jerome Gorski, University of Michigan), and screening cDNA libraries with cosmid probes (Sue Rider, ICRF).
The use of PCR on pools of YACs was discussed for mapping random end-sequenced cDNA clones to the physical map (Sikela). William Nierman (American Type Culture Collection) reported that 59% of PCR primers from end sequences were usable for mapping. Donald Moir (Collaborative Research, Inc.) presented the hybridization of cDNAs to gridded arrays of mega-YAC clones as an alternative to commonly used PCR-based cDNA mapping. David Beier (Harvard Medical School) demonstrated mapping mouse cDNAs by single-stranded conformation polymorphisms in inbred mouse strains.
Several groups are assembling transcriptional maps of whole chromosomes or chromosome regions: 21 (Katheleen Gardiner, Eleanor Roosevelt Institute; Yaspo; Andrew Peterson, University of California at San Francisco); Xq28 (Bernhard Korn, German Cancer Research Center; Toniolo; Hubert Smeets, University of Nijmegen); Xp21 (Francoise Muscatelli, ICRF); Xp11.21 (Gorski); Xq13.3 (Jozef Gecz, Slovak Academy of Science); and 7q22 (Johanna Rommens, Hospital for Sick Children). Different methods were applied singly or in combination. More data are needed to determine how effective the various methods will be over large regions and before gene distributions within chromosomal regions emerge.
One problem common to different methods of gene identification was how to define a gene. When can a hybrid-selected cDNA, an exon-trapped product, or a GRAIL-positive genomic fragment be identified positively as a coding sequence of a gene? Parameters suggested by workshop participants were reverse transcription PCR across an intron-exon boundary, signal on a Northern blot, and isolation of a poly Aþcontaining cDNA clone.
Another problem specific to candidate exons retrieved through exon trapping or hybrid selection is to avoid redundant analyses of multiple short fragments from the same gene. One possibility discussed is to isolate full-length cDNA clones with the first probe available and use this to rescreen the pool of other isolates (e.g., exon-trap products or hybrid-selected clones). This would identify other small cDNA clones in the selected library that are parts of the same gene.
Although each of these methods has advantages, no method is perfect. Future experiments will show which is best suited for a particular application, from cloning disease genes within a relatively small, defined region to assembling large-range transcriptional chromosome maps.
The Fourth International Transcribed Sequences Workshop is planned for October 16-18 in Montreal, Canada. [Contact: Nan Matthews; Eleanor Roosevelt Institute; 1899 Gaylord St.; Denver, CO 80206 (303/333-4515, Fax: -8423).]
[Ute Hochgeschwender, National Institute of Mental Health, and Katheleen Gardiner, Eleanor Roosevelt Institute]
The electronic form of the newsletter may be cited in the following style:
Human Genome Program, U.S. Department of Energy, Human Genome News (v5n6).
The Human Genome Project (HGP) was an international 13-year effort, 1990 to 2003. Primary goals were to discover the complete set of human genes and make them accessible for further biological study, and determine the complete sequence of DNA bases in the human genome. See Timeline for more HGP history.
Published from 1989 until 2002, this newsletter facilitated HGP communication, helped prevent duplication of research effort, and informed persons interested in genome research.