Human Genome Center 
Lawrence Berkeley National Laboratory 
1 Cyclotron Road 
Berkeley, CA 94720 

Mohandas Narla 
510/486-7029, Fax: -6746 
mohandas_narla@macmail.lbl.gov 

Joyce Pfeiffer 
Administrative Assistant 

Michael Palazzolo* 
Director, 1996­97 
 
*Now at Amgen, Inc. 
 

In lieu of individual abstracts, research projects and investigators at the LANL Center for Human Genome Studies are represented in this narrative. More information can be found on the center's Web site. 
 
 

DNA Prep Machine  (31k JPG)

Now, after another decade, an expansion of biological research relevant to Human Genome Project goals is being carried out within the Life Sciences Division, with support from the Information and Computing Sciences and Engineering divisions. Individuals in these research projects are making important new contributions to the key fields of molecular, cellular, and structural biology; physical chemistry; data management; and scientific instrumentation. Additionally, industry involvement in this growing venture is stimulated by Berkeley Lab's location in the San Francisco Bay area, home to the largest congregation of biotechnology research facilities in the world. 

In July 1997 the Berkeley genome center became part of the Joint Genome Institute. 

Sequencing 

Large-scale genomic sequencing has been a central, ongoing activity at Berkeley Lab since 1991. It has been funded jointly by DOE (for human genome production sequencing and technology development) and the NIH National Human Genome Research Institute [for sequencing the Drosophila melanogaster model system, which is carried out in partnership with the University of California, Berkeley (UCB)]. The human genome sequencing area at Berkeley Lab consists of five groups: Bioinstrumentation, Automation, Informatics, Biology, and Development. Complementing these activities is a group in Life Sciences Division devoted to functional genomics, including the transgenics program. 

The directed DNA sequencing strategy (see below, left) at Berkeley Lab was designed and implemented to increase the efficiency of genomic sequencing. A key element of the directed approach is maintaining information about the relative positions of potential sequencing templates throughout the entire sequencing process. Thus, intelligent choices can be made about which templates to sequence, and the number of selected templates can be kept to a minimum. More important, knowledge of the interrelationship of sequencing runs guides the assembly process, making it more resistant to difficulties imposed by repeated sequences. As of July 3, 1997, Berkeley Lab had generated 4.4 megabases of human sequence and, in collaboration with UCB, had tallied 7.6 megabases of Drosophila sequence. 

Instrumentation and Automation 
 
The instrumentation and automation program encompasses the design and fabrication of custom apparatus to facilitate experiments, the programming of laboratory robots to automate repetitive procedures, and the development of (1) improved hardware to extend the applicability range of existing commercial robots and (2) an integrated operating system to control and monitor experiments. Although some discrete instrumentation modules used in the integrated protocols are obtained commercially, LBNL designs its own custom instruments when existing capabilities are inadequate. The instrumentation modules are then integrated into a large system to facilitate large-scale production sequencing. In addition, a significant effort is devoted to improving fluorescence-assay methods, including DNA sequence analysis and mass spectrometry for molecular sizing. 

Recent advances in the instrumentation group include DNA Prep machine and Prep Track. These instruments are designed to automate completely the highly repetitive and labor-intensive DNA-preparation procedure to provide higher daily throughput and DNA of consistent quality for sequencing (see photo below and at left). 

Berkeley Lab's near-term needs are for 960 samples per day of DNA extracted from overnight bacteria growths. The DNA protocol is a modified boil prep prepared in a 96-well format. Overnight bacteria growths are lysed, and samples are separated from cell debris by centrifugation. The DNA is recovered by ethanol precipitation. 
 

The informatics group is focused on hardware and software support and system administration, software development for end sequencing, transposon mapping and sequence template selection, data-flow automation, gene finding, and sequence analysis. Data-flow automation is the main emphasis. Six key steps have been identified in this process, and software is being written and tested to automate all six. The first step involves controlling gel quality, trimming vector sequence, and storing the sequences in a database. A program module called Move-Track-Trim, which is now used in production, was written to handle these steps. The second through fourth steps in this process involve assembling, editing, and reconstructing P1 clones of 80,000 base pairs from 400-base traces. The fifth step is sequence annotation, and the sixth is data submission. 

Annotation can greatly enhance the biological value of these sequences. Useful annotations include homologies to known genes, possible gene locations, and gene signals such as promoters. LBNL is developing a workbench for automatic sequence annotation and annotation viewing and editing. The goal is to run a series of sequence-analysis tools and display the results to compare the various predictions. Researchers then will be able to examine all the annotations (for example, genes predicted by various gene-finding methods) and select the ones that look best. 

Nomi Harris developed Genotator, an annotation workbench consisting of a stand-alone annotation browser and several sequence-analysis functions. The back end runs several gene finders, homology searches (using BLAST), and signal searches and saves the results in ".ace" format. Genotator thus automates the tedious process of operating a dozen different sequence-analysis programs with many different input and output formats. Genotator can function via command-line arguments or with the graphical user interface. 

Chromosome 5 
Over the last year, the center has focused its production genomic sequencing on the distal 40 megabases of the human chromosome 5 long arm. This region was chosen because it contains a cluster of growth factor and receptor genes and is likely to yield new and functionally related genes through long-range sequence analysis. Results to date include: 

· 40-megabase nonchimeric map containing 82 yeast artificial chromosomes (YACs) in the chromosome 5 distal long arm. 

· 20-megabase contig map in the region of 5q23-q33 that contains 198 P1s, 60 P1 artificial chromosomes, and 495 bacterial artificial chromosomes (BACs) linked by 563 sequenced tagged sites (STSs) to form contigs. 

· 20-megabase bins containing 370 BACs in 74 bins in the region of 5q33-q35. 

Chromosome 21 
An early project in the study of Down syndrome (DS), which is characterized by chromosome 21 trisomy, constructed a high-resolution clone map in the chromosome 21 DS region to be used as a pilot study in generating a contiguous gene map for all of chromosome 21. This project has integrated P1 mapping efforts with transgenic studies in the Life Sciences Division. P1 maps provide a suitable form of genomic DNA for isolating and mapping cDNA. 

· 186 clones isolated in the major DS region of chromosome 21 comprising about 3megabases of genomic DNA extending from D21S17 to ETS2. Through cross-hybridization, overlapping P1s were identified, as well as gaps between two P1 contigs, and transgenic mice were created from P1 clones in the DS region for use in phenotypic studies. 

Transgenic Mice 
One of the approaches for determining the biological function of newly identified genes uses YAC transgenic mice. Human sequence harbored by YACs in transgenic mice has been shown to be correctly regulated both temporally and spatially. A set of nonchimeric overlapping YACs identified from the 5q31 region has been used to create transgenic mice. This set of transgenic mice, which together harbor 1.5 megabases of human sequence, will be used to assess the expression pattern and potential function of putative genes discovered in the 5q31 region. Additional mapping and sequencing are under way in a region of human chromosome 20 amplified in certain breast tumor cell lines. 

Resource for Molecular Cytogenetics 
Divining landmarks for human disease amid the enormous plain of the human genetic map is the mission of an ambitious partnership among the Berkeley Lab; University of California, San Francisco; and a diagnostics company. The collaborative Resource for Molecular Cytogenetics is charting a course toward important sites of biological interest on the 23 pairs of human chromosomes. 

The Resource employs the many tools of molecular cytogenetics. The most basic of these tools, and the cornerstone of the Resource's portfolio of proprietary technology, is a method generally known as "chromosome painting," which uses a technique referred to as fluorescence in situ hybridization or FISH. This technology was invented by LBNL Resource leaders Joe Gray and Dan Pinkel. 

A technology to emerge recently from the Resource is known as "Quantitative DNA Fiber Mapping (QDFM)." High-resolution human genome maps in a form suitable for DNA sequencing traditionally have been constructed various methods of fingerprinting, hybridization, and identification of overlapping STSs. However, these techniques do not readily yield information about sequence orientation, the extent of overlap of these elements, or the size of gaps in the map. Ulli Weier of the Resource developed the QDFM method of physical map assembly that enables the mapping of cloned DNA directly onto linear, fully extended DNA molecules. QDFM allows unambiguous assembly of critical elements leading to high-resolution physical maps. This task now can be accomplished in less than 2days, as compared with weeks by conventional methods. QDFM also enables detection and characterization of gaps in existing physical mapsa crucial step toward completing a definitive human genome map.