The Burrows-Wheeler transform (BWT) of short-read data has unexplored potential utilities, such as for efficient and sensitive variation analysis against multiple reference genome sequences, because it does not depend on any particular reference genome sequence, unlike conventional mapping-based methods. However, since the amount of read data is generally much larger than the size of the reference sequence, computation of the BWT of reads is not easy, and this hampers development of potential applications. For the alleviation of this problem, a new method of computing the BWT of reads in parallel is proposed. The BWT, corresponding to a sorted list of suffixes of reads, is constructed incrementally by successively including longer and longer suffixes. The working data is divided into more than 10,000 "blocks" corresponding to sublists of suffixes with the same prefixes. Thousands of groups of blocks can be processed in parallel while making exclusive writes and concurrent reads into a shared memory. Reads and writes are basically sequential, and the read concurrency is limited to two. Thus, a fine-grained parallelism, referred to as prefix parallelism, is expected to work efficiently. The time complexity for processing n reads of length l is O(nl(2)). On actual biological DNA sequence data of about 100 Gbp with a read length of 100 bp (base pairs), a tentative implementation of the proposed method took less than an hour on a single-node computer; i.e., it was about three times faster than one of the fastest programs developed so far.

MOTIVATION:Sequence-variation analysis is conventionally performed on mapping results that are highly redundant and occasionally contain undesirable heuristic biases. A straightforward approach to single-nucleotide polymorphism (SNP) analysis, using the Burrows-Wheeler transform (BWT) of short-read data, is proposed.RESULTS:The BWT makes it possible to simultaneously process collections of read fragments of the same sequences; accordingly, SNPs were found from the BWT much faster than from the mapping results. It took only a few minutes to find SNPs from the BWT (with a supplementary data, fragment depth of coverage [FDC]) using a desktop workstation in the case of human exome or transcriptome sequencing data and 20 min using a dual-CPU server in the case of human genome sequencing data. The SNPs found with the proposed method almost agreed with those found by a time-consuming state-of-the-art tool, except for the cases in which the use of fragments of reads led to sensitivity loss or sequencing depth was not sufficient. These exceptions were predictable in advance on the basis of minimum length for uniqueness (MLU) and FDC defined on the reference genome. Moreover, BWT and FDC were computed in less time than it took to get the mapping results, provided that the data were large enough.

BACKGROUND:The potential utility of the Burrows-Wheeler transform (BWT) of a large amount of short-read data ("reads") has not been fully studied. The BWT basically serves as a lossless dictionary of reads, unlike the heuristic and lossy reads-to-genome mapping results conventionally obtained in the first step of sequence analysis. Thus, it is naturally expected to lead to development of sensitive methods for analysis of short-read data. Recently, one of the most active areas of research in sequence analysis is sensitive detection of rare genomic rearrangements from whole-genome sequencing (WGS) data of heterogeneous cancer samples. The application the BWT of reads to the analysis of genomic rearrangements is addressed in this study.RESULTS:A new method for sensitive detection of genomic rearrangements by using the BWT of reads in the following three steps is proposed: first, breakpoint regions, which contain breakpoints and are joined together by rearrangement, are predicted from the distribution of so-called discordant pairs by using a kind of the conjugate gradient method; second, reads partially matching the breakpoint regions are collected from the BWT of reads; and third, breakpoints are detected as branching points among the collected reads, and their precise positions are determined. The method was experimentally implemented, and its performance (i.e., sensitivity and specificity) was evaluated by using simulated data with known artificial rearrangements. It was applied to publicly available real biological WGS data of cancer patients, and the detection results were compared with published results.CONCLUSIONS:Serving as a lossless dictionary of reads, the BWT of short reads enables sensitive analysis of genomic rearrangements in heterogeneous cancer-genome samples when used in conjunction with breakpoint-region predictions based on a conjugate gradient method.

Myers' elegant and powerful bit-parallel dynamic programming algorithm for approximate string matching has a restriction that the query length should be within the word size of the computer, typically 64. We propose a modification of Myers' algorithm, in which the modification has a restriction not on the query length but on the maximum number of mismatches (substitutions, insertions, or deletions), which should be less than half of the word size. The time complexity is O(m log |Σ|), where m is the query length and |Σ| is the size of the alphabet Σ. Thus, it is particularly suited for sequences on a small alphabet such as DNA sequences. In particular, it is useful in quickly extending a large number of seed alignments against a reference genome for high-throughput short-read data produced by next-generation DNA sequencers.

Spaced seeds have been shown to be superior to continuous seeds for efficient and sensitive homology search based on the seed-and-extend paradigm. Much the same is true in genome mapping of high-throughput short-read data. However, a highly sensitive search with multiple spaced patterns often requires the use of a great amount of index data. We propose a novel seed-set construction method for efficient and sensitive genome mapping of short reads with relatively high error rates, which uses only continuous seeds of variable length allowing a few errors. The seed lengths and allowable error positions are optimized on the basis of entropy, which is a measure of ambiguity or repetitiveness of mapping positions. These seeds can be searched efficiently using the Burrows-Wheeler transform of the reference genome. Evaluation using actual biological SOLiD sequence data demonstrated that our method was competitive in speed and sensitivity using much less memory and disk space in comparison to spaced-seed methods.

Abstract We have developed efficient in-practice algorithms for computing rank and select functions on a binary string, based on a novel data structure, a hierarchical binary string with hierarchical accumulatives. It efficiently stores decomposed information on partial summations over various scales of subregions of a given binary string, so that the required space overhead ratio is only about 3.5% irrespective of the string length. Values of rank and select functions are computed hierarchically in [(log(2)n)/8] iterations, where n is the string length. For example, for an unbiased random binary string of 64 G bits, each value of these functions can be computed in about a microsecond, on average, on a single 3.0-GHz CPU using 8+ GB of memory. We also present their applications to genome mapping problems for large-scale short-read DNA sequence data, especially produced by ultra-high-throughput new-generation DNA sequencers. The algorithms are applied to the binarization of the Burrows-Wheeler transform of the human genome DNA sequence. For the sake of high-speed performance, we adopted a somewhat stringent mapping condition that allows at most a single-base mismatch (either a substitution, insertion, or deletion of a single base) per query sequence. An experimentally implemented program mapped several thousands of sequences per second on a single 3.0-GHz CPU, several times faster than ELAND, a widely used mapping program with the Illumina-Solexa 1G analyser.

We analyzed diversity of mRNA produced as a result of alternative splicing in order to evaluate gene function. First, we predicted the number of human genes transcribed into protein-coding mRNAs by using the sequence information of full-length cDNAs and 5'-ESTs and obtained 23 241 of such human genes. Next, using these genes, we analyzed the mRNA diversity and consequently sequenced and identified 11 769 human full-length cDNAs whose predicted open reading frames were different from other known full-length cDNAs. Especially, 30% of the cDNAs we identified contained variation in the transcription start site (TSS). Our analysis, which particularly focused on multiple variable first exons (FEVs) formed due to the alternative utilization of TSSs, led to the identification of 261 FEVs expressed in the tissue-specific manner. Quantification of the expression profiles of 13 genes by real-time PCR analysis further confirmed the tissue-specific expression of FEVs, e.g. OXR1 had specific TSS in brain and tumor tissues, and so on. Finally, based on the results of our mRNA diversity analysis, we have created the FLJ Human cDNA Database. From our result, it has been understood mechanisms that one gene produces suitable protein-coding transcripts responding to the situation and the environment.

Completion of human genome sequencing has greatly accelerated functional genomic research. Full-length cDNA clones are essential experimental tools for functional analysis of human genes. In one of the projects of the New Energy and Industrial Technology Development Organization (NEDO) in Japan, the full-length human cDNA sequencing project (FLJ project), nucleotide sequences of approximately 30 000 human cDNA clones have been analyzed. The Gateway system is a versatile framework to construct a variety of expression clones for various experiments. We have constructed 33 275 human Gateway entry clones from full-length cDNAs, representing to our knowledge the largest collection in the world. Utilizing these clones with a highly efficient cell-free protein synthesis system based on wheat germ extract, we have systematically and comprehensively produced and analyzed human proteins in vitro. Sequence information for both amino acids and nucleotides of open reading frames of cDNAs cloned into Gateway entry clones and in vitro expression data using those clones can be retrieved from the Human Gene and Protein Database (HGPD, http://www.HGPD.jp). HGPD is a unique database that stores the information of a set of human Gateway entry clones and protein expression data and helps the user to search the Gateway entry clones.

We introduce a new data structure, a localized suffix array, based on which occurrence information is dynamically represented as the combination of global positional information and local lexicographic order information in text search applications. For the search of a pair of words within a given distance, many candidate positions that share a coarse-grained global position can be compactly represented in term of local lexicographic orders as in the conventional suffix array, and they can be simultaneously examined for violation of the distance constraint at the coarse-grained resolution. Trade-off between the positional and lexicographical information is progressively shifted towards finer positional resolution, and the distance constraint is reexamined accordingly. Thus the paired search can be efficiently performed even if there are a large number of occurrences for each word. The localized suffix array itself is in fact a reordering of bits inside the conventional suffix array, and their memory requirements are essentially the same. We demonstrate an application to genome mapping problems for paired-end short reads generated by new-generation DNA sequencers. When paired reads are highly repetitive, it is time-consuming to naïvely calculate, sort, and compare all of the coordinates. For a human genome re-sequencing data of 36 base pairs, more than 10 times speedups over the naïve method were observed in almost half of the cases where the sums of redundancies (number of individual occurrences) of paired reads were greater than 2,000.

Appropriate resources and expression technology necessary for human proteomics on a whole-proteome scale are being developed. We prepared a foundation for simple and efficient production of human proteins using the versatile Gateway vector system. We generated 33,275 human Gateway entry clones for protein synthesis, developed mRNA expression protocols for them and improved the wheat germ cell-free protein synthesis system. We applied this protein expression system to the in vitro expression of 13,364 human proteins and assessed their biological activity in two functional categories. Of the 75 tested phosphatases, 58 (77%) showed biological activity. Several cytokines containing disulfide bonds were produced in an active form in a nonreducing wheat germ cell-free expression system. We also manufactured protein microarrays by direct printing of unpurified in vitro-synthesized proteins and demonstrated their utility. Our 'human protein factory' infrastructure includes the resources and expression technology for in vitro proteome research.

Here we report the new features and improvements in our latest release of the H-Invitational Database (H-InvDB; http://www.h-invitational.jp/), a comprehensive annotation resource for human genes and transcripts. H-InvDB, originally developed as an integrated database of the human transcriptome based on extensive annotation of large sets of full-length cDNA (FLcDNA) clones, now provides annotation for 120 558 human mRNAs extracted from the International Nucleotide Sequence Databases (INSD), in addition to 54 978 human FLcDNAs, in the latest release H-InvDB_4.6. We mapped those human transcripts onto the human genome sequences (NCBI build 36.1) and determined 34 699 human gene clusters, which could define 34 057 (98.1%) protein-coding and 642 (1.9%) non-protein-coding loci; 858 (2.5%) transcribed loci overlapped with predicted pseudogenes. For all these transcripts and genes, we provide comprehensive annotation including gene structures, gene functions, alternative splicing variants, functional non-protein-coding RNAs, functional domains, predicted sub cellular localizations, metabolic pathways, predictions of protein 3D structure, mapping of SNPs and microsatellite repeat motifs, co-localization with orphan diseases, gene expression profiles, orthologous genes, protein–protein interactions (PPI) and annotation for gene families. The current H-InvDB annotation resources consist of two main views: Transcript view and Locus view and eight sub-databases: the DiseaseInfo Viewer, H-ANGEL, the Clustering Viewer, G-integra, the TOPO Viewer, Evola, the PPI view and the Gene family/group.

We propose a novel design and fabrication technique for a micro direct methanol fuel cell (μ-Direct methanol fuel cells (DMFC)). Ten μ-DMFCs were fabricated on a polymeric flexible substrate containing a micro-holes array (∅=500μm). Our concept is that a high voltage DMFC can be achieved through the integration of a micro DMFC array on one flexible substrate. With 10 cells in series on the substrate, we achieved an open circuit voltage of 5.6V and an average cell power density of 3mWcm−2. Further, none of the cells failed after subjecting the substrate to 1000 fold repetitions. Preliminary test results show that our DMFC fabricated on a polymeric substrate is useful as a miniature DMFC.

By analyzing 1,780,295 5'-end sequences of human full-length cDNAs derived from 164 kinds of oligo-cap cDNA libraries, we identified 269,774 independent positions of transcriptional start sites (TSSs) for 14,628 human RefSeq genes. These TSSs were clustered into 30,964 clusters that were separated from each other by more than 500 bp and thus are very likely to constitute mutually distinct alternative promoters. To our surprise, at least 7674 (52%) human RefSeq genes were subject to regulation by putative alternative promoters (PAPs). On average, there were 3.1 PAPs per gene, with the composition of one CpG-island-containing promoter per 2.6 CpG-less promoters. In 17% of the PAP-containing loci, tissue-specific use of the PAPs was observed. The richest tissue sources of the tissue-specific PAPs were testis and brain. It was also intriguing that the PAP-containing promoters were enriched in the genes encoding signal transduction-related proteins and were rarer in the genes encoding extracellular proteins, possibly reflecting the varied functional requirement for and the restricted expression of those categories of genes, respectively. The patterns of the first exons were highly diverse as well. On average, there were 7.7 different splicing types of first exons per locus partly produced by the PAPs, suggesting that a wide variety of transcripts can be achieved by this mechanism. Our findings suggest that use of alternate promoters and consequent alternative use of first exons should play a pivotal role in generating the complexity required for the highly elaborated molecular systems in humans.

We have developed an in silico method of selection of human full-length cDNAs encoding secretion or membrane proteins from oligo-capped cDNA libraries. Fullness rates were increased to about 80% by combination of the oligo-capping method and ATGpr, software for prediction of translation start point and the coding potential. Then, using 5'-end single-pass sequences, cDNAs having the signal sequence were selected by PSORT ('signal sequence trap). We also applied 'secretion or membrane protein-related keyword trap' based on the result of BLAST search against the SWISS-PROT database for the cDNAs which could not be selected by PSORT. Using the above procedures, 789 cDNAs were primarily selected and subjected to full-length sequencing, and 334 of these cDNAs were finally selected as novel. Most of the cDNAs (295 cDNAs: 88.3%) were predicted to encode secretion or membrane proteins. In particular, 165 (80.5%) of the 205 cDNAs selected by PSORT were predicted to have signal sequences, while 70 (54.2%) of the 129 cDNAs selected by 'keyword trap' preserved the secretion or membrane protein-related keywords. Many important cDNAs were obtained, including transporters, receptors, and ligands, involved in significant cellular functions. Thus, an efficient method of selecting secretion or membrane protein-encoding cDNAs was developed by combining the above four procedures.

The human genome sequence defines our inherent biological potential; the realization of the biology encoded therein requires knowledge of the function of each gene. Currently, our knowledge in this area is still limited. Several lines of investigation have been used to elucidate the structure and function of the genes in the human genome. Even so, gene prediction remains a difficult task, as the varieties of transcripts of a gene may vary to a great extent. We thus performed an exhaustive integrative characterization of 41,118 full-length cDNAs that capture the gene transcripts as complete functional cassettes, providing an unequivocal report of structural and functional diversity at the gene level. Our international collaboration has validated 21,037 human gene candidates by analysis of high-quality full-length cDNA clones through curation using unified criteria. This led to the identification of 5,155 new gene candidates. It also manifested the most reliable way to control the quality of the cDNA clones. We have developed a human gene database, called the H-Invitational Database (H-InvDB; http://www.h-invitational.jp/). It provides the following: integrative annotation of human genes, description of gene structures, details of novel alternative splicing isoforms, non-protein-coding RNAs, functional domains, subcellular localizations, metabolic pathways, predictions of protein three-dimensional structure, mapping of known single nucleotide polymorphisms (SNPs), identification of polymorphic microsatellite repeats within human genes, and comparative results with mouse full-length cDNAs. The H-InvDB analysis has shown that up to 4% of the human genome sequence (National Center for Biotechnology Information build 34 assembly) may contain misassembled or missing regions. We found that 6.5% of the human gene candidates (1,377 loci) did not have a good protein-coding open reading frame, of which 296 loci are strong candidates for non-protein-coding RNA genes. In addition, among 72,027 uniquely mapped SNPs and insertions/deletions localized within human genes, 13,215 nonsynonymous SNPs, 315 nonsense SNPs, and 452 indels occurred in coding regions. Together with 25 polymorphic microsatellite repeats present in coding regions, they may alter protein structure, causing phenotypic effects or resulting in disease. The H-InvDB platform represents a substantial contribution to resources needed for the exploration of human biology and pathology.

At Visual Arts College Osaka, we are exploring ways to use the computer as an instrument for expressing art. Usually, our students learn to create works of Web, CD-ROM and video content using the computer. We had been thinking about using Squeak for a while, however, it was not easy to introduce Squeak in our classes because of the issues relevant to the need of using the Japanese language in Squeak and the need of Japanese manuals. Fortunately, we met with Mr. Kazuhiro Abe in May 2002, who was developing a Japanese edition of Squeak. We decided to use Squeak in our class for 3 months aiming at making Squeak "our instruments". This attempt was experimental.

1 Central Research Laboratory, Hitachi Ltd., 1-280 Higashi-Koigakubo, Kokubunji, Tokyo 185-8601, Japan 2 Human Genome Center, Institute of Medical Science, University of Tokyo, 4-6-1 Shirokane-dai, Minato-ku, Tokyo 108-8639, Japan 3 Japan Biological Information Research Center, National Institute of Advanced Industrial Science and Technology, 2-41-6 Aomi, Koto-ku, Tokyo 135-0064, Japan 4 Reverse Proteomics Research Institute, 1532-3 Yana, Kisarazu, Chiba, Japan