The US DOE Joint Genome Institute (JGI) is a high-throughput sequencing center involved in a myriad of sequencing projects. A major effort at JGI is the sequencing of microbial genomes of relevance to the DOE missions of carbon sequestration, bioremediation and energy production. The JGI Microbial Program is responsible for the generation of over 200 microbial genomes and we are interested in utilizing new technologies to increase capacity. The 454 sequencing platform is an integrated system of emulsion-based PCR amplification of hundreds of thousands of DNA fragments linked to high throughput parallel pyrosequencing in picoliter-sized wells. The 454 sequencing platform can deliver over 20 million base pairs (mbp) from a single run, however, the quality of resulting assembly has not been well characterized. We sequenced Prochlorococcus marinus NATL2A (DNA kindly provided by Penny Chisolm and Claudia Steglich, Department of Biology, MIT) with one run on 454 platform, and generated over 36 mbp of data. The 332,387 reads of average length 109 bp were assembled with 454 s Newbler assembly tool which generated contigs of consensus sequence. We compared the assembly results from the 454 platform with sequence previously finished at JGI using traditional Sanger sequencing technology. The methods were compared for error rates, misassemblies, coverage and depth. We will also discuss the feasibility of using the 454 platform for de novo whole genome shotgun sequencing and the optimal ratio of data from 454 and Sanger sequencing to achieve high quality finished genome sequences in a time and cost effective manner.

The US DOE Joint Genome Institute (JGI) is a high-throughput sequencing center involved in a myriad of sequencing projects. A major effort at JGI is the sequencing of microbial genomes of relevance to the DOE missions of carbon sequestration, bioremediation and energy production. The JGI Microbial Program is responsible for the generation of over 200 microbial genomes and we are interested in utilizing new technologies to increase capacity. The 454 sequencing platform is an integrated system of emulsion-based PCR amplification of hundreds of thousands of DNA fragments linked to high throughput parallel pyrosequencing in picoliter-sized wells. The 454 sequencing platform can deliver 30 to 50 million base pairs (mbp) from a single run, however, our previous study revealed that the quality of the resulting assembly contains high numbers of misassemblies and base errors due to short read length and lack of paired-end information. The traditional Sanger sequencing method is lower in throughput and more costly but it provides high quality sequencing results and more accurate assemblies. The paired-end information from Sanger sequencing is proven to be crucial in scaffolding and gap closure. We combined 454 sequencing results with different amounts of paired Sanger sequencing reads from three different sized shotgun libraries and analyzed the results. Assemblies from all combinations were done by Newbler and Phred/Phrap and viewed and analyzed by Consed and in-house developed software. Numbers of remaining gaps and low quality regions from different combinations were assessed. Distribution of coverage and possible errors were analyzed for both platforms. We will also discuss the optimal ratio of data from 454 and Sanger sequencing to achieve high quality finished microbial genome sequences in a time and cost effective manner.

At JGI, major efforts have been spent on sequencing through gaps in the assemblies generated by whole genome shotgun sequencing. Traditional shotgun sequencing is known to have difficulty in both cloning of A/T rich regions and sequencing of G/C rich regions. To help alleviate this problem we have applied the 454 sequencing platform as another tool for gap closure. Although 454 sequencing has been shown to have difficulty with homopolymer, it does not have the same biases as traditional sequencing. Therefore these two approaches together can be complementary. Different strategies have been tested for applying 454 sequencing in gap closure stage of WGS sequencing. Direct shotgun sequencing using 454 platform has been combined with traditional Sanger sequencing at the final assembly stage. We also developed a protocol in which gap-spanning clones are pooled, sequenced and assembled with 454 platform and the resulting contigs are added into their respective Sanger assemblies. The quality of the final assemblies from these different strategies was examined and compared. The regions only covered by 454 sequencing were studied and the sequencing features of these regions were analyzed. The base quality and assembly correctness of these regions were also assessed. Detailed results will be presented.

The US DOE Joint Genome Institute (JGI) is a high-throughput genomics facility involved in sequencing a variety of organisms. A major effort at JGI is the sequencing of genomes and microbial community samples of relevance to the DOE missions of carbon sequestration, bioremediation and energy production. cDNA/EST sequencing is an integral part of genomic sequencing because it provides crucial information for gene models and genome annotation. The 454 sequencing platform is an integrated system of emulsion-based PCR amplification of hundreds of thousands of DNA fragments linked to high throughput parallel pyrosequencing in picoliter-sized wells. Several strategies have been designed and carried out at JGI to use the 454 platform for cDNA/EST sequencing. cDNA libraries constructed by conventional methods were subjected to direct 454 sequencing. In addition, special primers and adaptors were also designed for library construction so the directional sequencing feature of the 454 technology can be used to sequence a particular end of the cDNA/EST fragments. Adaptor sequences used by 454 library construction can be incorporated into polyT primer, cap primer and/or random primer for cDNA/EST library construction. The 454 sequencing platform can deliver 200 to 400 thousand cDNA/EST reads from a single run and does not require cloning step, potentially improving the coverage obtained through traditional Sanger sequencing. The large numbers of short reads generated by the 454 platform can be aligned to genome assemblies to extend and confirm gene models. Results from different strategies of library construction combined with 454 sequencing will be presented. The coverage of the library and the novelty rate are compared with traditional Sanger sequencing. The possible assembly problems caused by short reads with slightly higher error rate from 454 will also be addressed.

The Department of Energy Joint Genome Institute (www.jgi.doe.gov) in Walnut Creek, CA is a high throughput DNA sequencing facility with a current throughput of approximately 3 billion Sanger base pairs per month. A major effort at JGI is the sequencing of microbial genomes of relevance to the DOE missions of carbon sequestration, bioremediation and energy production. The JGI Microbial Program and Community Sequencing Program together are responsible for the generation of sequencing data for over 400 microbial genomes. At the traditional Sanger sequencing side, JGI is running about 70 ABI sequencers on a 24/7 schedule and about 40 GE MegaBACE 4500 sequencers on a 24/5 schedule. JGI currently runs 2 Roche's GS20 instruments to supplement our traditional Sanger sequencing. Our current whole genome shotgun sequencing strategy is to sequence 3kb and/or 8kb shotgun libraries to a combined 4-8x draft coverage and to sequence fosmid ends to 1x sequence coverage with Sanger sequencing and to supplement that with 12-25x coverage with 454 sequencing platform depending on the sizes of the genomes. For new microbial genomes we initiated, 454 sequencing was carried out at the same time the shotgun cloning for Sanger sequencing started. 454 sequencing data was used to profile the genomes for G/C content, genome sizes and other features of the genomes. For existing microbial genome projects for which the Sanger sequencing data has already been generated, we have been adding 454 sequencing coverage at the finishing stage. 454 sequencing data was assembled by default Newbler assembler software package from 454 Life Sciences. The Newbler contigs were then fragmented and the quality and coverage information of the contigs was captured by in-house developed software tool packages. The fragmentation strategy we currently use is to cut the Newbler contigs into 750 bp fragments with 100 bp overlap. The overlapping fragments from Newbler contigs were finally assembled with Sanger sequencing data using the assembler(s) of our choice. The gaps and low quality areas in the final assembly were manually sequenced to JGI defined quality standard. At this point, the genome is ready for analysis and annotation. 454 sequencing technology is also used more directly in gap closure stage of the microbial genome whole genome shotgun sequencing. Gap spanning clones from multiple genomes were pooled together and the resulting DNA was subjected to 454 sequencing. The Newbler assembly results from pooled clone sequencing were added to final genome assemblies to fill the gaps.

We previously reported MetaBAT, an automated metagenome binning software tool to reconstruct single genomes from microbial communities for subsequent analyses of uncultivated microbial species. MetaBAT has become one of the most popular binning tools largely due to its computational efficiency and ease of use, especially in binning experiments with a large number of samples and a large assembly. MetaBAT requires users to choose parameters to fine-tune its sensitivity and specificity. If those parameters are not chosen properly, binning accuracy can suffer, especially on assemblies of poor quality. Here we developed MetaBAT 2 to overcome this problem. MetaBAT 2 uses a new adaptive binning algorithm to eliminate manual parameter tuning. We also performed extensive software engineering optimization to increase both computational and memory efficiency. Comparing MetaBAT 2 to alternative software tools on over 100 real world metagenome assemblies shows superior accuracy and computing speed. Binning a typical metagenome assembly takes only a few minutes on a single commodity workstation. We therefore recommend the community adopts MetaBAT 2 for their metagenome binning experiments. MetaBAT 2 is open source software and available at https://bitbucket.org/berkeleylab/metabat .

We previously reported on MetaBAT, an automated metagenome binning software tool to reconstruct single genomes from microbial communities for subsequent analyses of uncultivated microbial species. MetaBAT has become one of the most popular binning tools largely due to its computational efficiency and ease of use, especially in binning experiments with a large number of samples and a large assembly. MetaBAT requires users to choose parameters to fine-tune its sensitivity and specificity. If those parameters are not chosen properly, binning accuracy can suffer, especially on assemblies of poor quality. Here, we developed MetaBAT 2 to overcome this problem. MetaBAT 2 uses a new adaptive binning algorithm to eliminate manual parameter tuning. We also performed extensive software engineering optimization to increase both computational and memory efficiency. Comparing MetaBAT 2 to alternative software tools on over 100 real world metagenome assemblies shows superior accuracy and computing speed. Binning a typical metagenome assembly takes only a few minutes on a single commodity workstation. We therefore recommend the community adopts MetaBAT 2 for their metagenome binning experiments. MetaBAT 2 is open source software and available at https://bitbucket.org/berkeleylab/metabat.