|
|
|
Joint Genome Institute Achievements, Future Plans Presented A contingent of researchers from JGI reported on the challenges and triumphs involved in merging the sequencing capabilities of three national laboratories into a single, smoothly functioning unit while meeting much higher sequencing goals. Efforts were highly successful, and all objectives were reached as planned, with over 20 Mb of sequence submitted to GenBank in October 1998. JGI converged the disparate sequencing processes of the three centers of Lawrence Berkeley, Lawrence Livermore, and Los Alamos national laboratories at the Production Sequencing Facility (PSF) in Walnut Creek, California. The team took up residence in one of the new buildings in December 1998, and Secretary of Energy Bill Richardson was keynote speaker at the April 19 formal PSF dedication (see box at right). Clone Resources Jan-Fang Cheng reported that about 71.7 Mb of sequence-ready maps were generated in FY 1998 for target regions of chromosomes 5, 16, and 19, and some 3500 unique genes have been mapped to these regions. The goal for FY 1999 was to develop maps for over 250 Mb to allow production of a draft sequence of the three chromosomes by March 2000. The JGI mapping teams then plan to generate similar maps for the mouse genome in regions syntenic with these human chromosomes. Clones containing target regions are isolated from 10x-coverage BAC libraries via a combination of colony hybridization and PCR approaches using STSs obtained mostly from public databases. Minimally overlapping clones that represent contiguous regions (contigs) of the chromosome are selected for sequencing. Contigs, expanded by end-sequence STS walking, are oriented by STSs developed from known genes and ordered genetic and RH markers. All clones are sized by pulsed-field gel electrophoresis, and their chromosome map locations are confirmed by FISH. Detailed information on STS and restriction maps is available on the Web site. Quality Control and Assurance Norman Doggett described the robust system of quality control and assessment processes for sequence generated at JGI. He emphasized JGIs commitment to quality, with more than 95% of all finished bases having Phrap scores over 40 and at least 95% of all bases covered in reads from both strands (or 2 chemistries). To ensure correct assembly, JGI validates the data before submitting the sequence to GenBank and performs postsubmission quality assessment. Validation includes visual inspection of all bases with Phrap values of 30 or less and comparison of the final assembled sequence to 3 independent high-resolution restriction fingerprints. Informatics Meeting JGIs ambitious FY 1998 goals was something of a miracle of elasticity of the preexisting separate systems at member laboratories, according to Tom Slezak. He noted the complex efforts now under way as the laboratories work toward merging into a single PSF informatics system without interfering with production pressures. The long-term goal is to make information available in a single view through the entire process, from mapping to submission, while allowing for multiple viewing methods. Data silos in various locations currently are hard to connect, but eventually physical maps will become sequence maps as all marker and annotation data are merged. Querying, navigating, and displaying data soon will require seamless links to information on function, structure, and experimental results. Slezak emphasized the importance of implementing an infrastructure for functional genomics informatics. |
加入最爱