Serge Batalov

BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources

Online gene annotation resources are indispensable for analysis of genomics data. However, the landscape of these online resources is highly fragmented, and scientists often visit dozens of these sites for each gene in a candidate gene list. Here, we introduce BioGPS http://biogps.gnf.org, a centralized gene portal for aggregating distributed gene annotation resources. Moreover, BioGPS embraces the principle of community intelligence, enabling any user to easily and directly contribute to the BioGPS platform.

Functional annotation of a full-length mouse cDNA collection

The RIKEN Mouse Gene Encyclopaedia Project, a systematic approach to determining the full coding potential of the mouse genome, involves collection and sequencing of full-length complementary DNAs and physical mapping of the corresponding genes to the mouse genome. We organized an international functional annotation meeting (FANTOM) to annotate the first 21,076 cDNAs to be analysed in this project. Here we describe the first RIKEN clone collection, which is one of the largest described for any organism. Analysis of these cDNAs extends known gene families and identifies new ones.The RIKEN Mouse Gene Encyclopaedia Project, a systematic approach to determining the full coding potential of the mouse genome, involves collection and sequencing of full-length complementary DNAs and physical mapping of the corresponding genes to the mouse genome. We organized an international functional annotation meeting (FANTOM) to annotate the first 21,076 cDNAs to be analysed in this project. Here we describe the first RIKEN clone collection, which is one of the largest described for any organism. Analysis of these cDNAs extends known gene families and identifies new ones.

Genome-Wide and Functional Annotation of Human E3 Ubiquitin Ligases Identifies MULAN, a Mitochondrial E3 that Regulates the Organelle's Dynamics and Signaling

Specificity of protein ubiquitylation is conferred by E3 ubiquitin (Ub) ligases. We have annotated ∼617 putative E3s and substrate-recognition subunits of E3 complexes encoded in the human genome. The limited knowledge of the function of members of the large E3 superfamily prompted us to generate genome-wide E3 cDNA and RNAi expression libraries designed for functional screening. An imaging-based screen using these libraries to identify E3s that regulate mitochondrial dynamics uncovered MULAN/FLJ12875, a RING finger protein whose ectopic expression and knockdown both interfered with mitochondrial trafficking and morphology. We found that MULAN is a mitochondrial protein – two transmembrane domains mediate its localization to the organelles outer membrane. MULAN is oriented such that its E3-active, C-terminal RING finger is exposed to the cytosol, where it has access to other components of the Ub system. Both an intact RING finger and the correct subcellular localization were required for regulation of mitochondrial dynamics, suggesting that MULANs downstream effectors are proteins that are either integral to, or associated with, mitochondria and that become modified with Ub. Interestingly, MULAN had previously been identified as an activator of NF-κB, thus providing a link between mitochondrial dynamics and mitochondria-to-nucleus signaling. These findings suggest the existence of a new, Ub-mediated mechanism responsible for integration of mitochondria into the cellular environment.

Identification of Modulators of TRAIL-Induced Apoptosis via RNAi-Based Phenotypic Screening

New opportunities in mammalian functional genomics are emerging through the combination of high throughput technology and methods that allow manipulation of gene expression in living cells. Here we describe the application of an RNAi-based forward genomics approach toward understanding the biology and mechanism of TRAIL-induced apoptosis. TRAIL is a TNF superfamily member that induces selective cytotoxicity of tumor cells when bound to its cognate receptors. In addition to detecting well-characterized genes in the apoptosis pathway, we uncover several modulators including DOBI, a gene required for progression of the apoptotic signal through the intrinsic mitochondrial cell death pathway, and MIRSA, a gene that acts to limit TRAIL-induced apoptosis. Moreover, our data suggest a role for MYC and the WNT pathway in maintaining susceptibility to TRAIL. Collectively, these observations offer several insights on how TRAIL mediates the selective killing of tumor cells and demonstrate the utility of large-scale RNAi screens in mammalian cells.

Genome-wide single-nucleotide polymorphism analysis defines haplotype patterns in mouse

The nature and organization of polymorphisms, or differences, between genomes of individuals are of great interest, because these variations can be associated with or even underlie phenotypic traits, including disease susceptibility. To gain insight into the genetic and evolutionary factors influencing such biological variation, we have examined the arrangement (haplotype) of single-nucleotide polymorphisms across the genomes of eight inbred strains of mice. These analyses define blocks of high or low diversity, often extending across tens of megabases that are delineated by abrupt transitions. These observations provide a striking contrast to the haplotype structure of the human genome.

Use of a dense single nucleotide polymorphism map for in silico mapping in the mouse.

Rapid expansion of available data, both phenotypic and genotypic, for multiple strains of mice has enabled the development of new methods to interrogate the mouse genome for functional genetic perturbations. In silico mapping provides an expedient way to associate the natural diversity of phenotypic traits with ancestrally inherited polymorphisms for the purpose of dissecting genetic traits. In mouse, the current single nucleotide polymorphism (SNP) data have lacked the density across the genome and coverage of enough strains to properly achieve this goal. To remedy this, 470,407 allele calls were produced for 10,990 evenly spaced SNP loci across 48 inbred mouse strains. Use of the SNP set with statistical models that considered unique patterns within blocks of three SNPs as an inferred haplotype could successfully map known single gene traits and a cloned quantitative trait gene. Application of this method to high-density lipoprotein and gallstone phenotypes reproduced previously characterized quantitative trait loci (QTL). The inferred haplotype data also facilitates the refinement of QTL regions such that candidate genes can be more easily identified and characterized as shown for adenylate cyclase 7.

Ubiquitin-mediated sequestration of normal cellular proteins into polyglutamine aggregates

A hallmark of most neurodegenerative diseases, including those caused by polyglutamine expansion, is the formation of ubiquitin (Ub)-positive protein aggregates in affected neurons. This finding suggests that the Ub system may be involved in common mechanisms underlying these otherwise unrelated diseases. Here we report the finding of ataxin-3 (Atx-3), whose mutation is implicated in the neurodegenerative disease spinocerebellar ataxia type 3, in a bioinformatics search of the human genome for components of the Ub system. We show that wild-type Atx-3 is a Ub-binding protein and that the interaction of Atx-3 with Ub is mediated by motifs homologous to those found in a proteasome subunit. Both wild-type Atx-3 and the otherwise unrelated Ub-binding protein p62/Sequestosome-1 have been shown to be sequestered into aggregates in affected neurons in several neurodegenerative diseases, but the mechanism for this recruitment has remained unclear. In this article, we show that functional Ub-binding motifs in Atx-3 and p62 proteins are required for the localization of both proteins into aggregates in a cell-based assay that recapitulates several features of polyglutamine disease. We propose that the Ub-mediated sequestration of essential Ub-binding protein(s) into aggregates may be a common mechanism contributing to the pathogenesis of neurodegenerative diseases.

A Comparison of the Celera and Ensembl Predicted Gene Sets Reveals Little Overlap in Novel Genes

genes, but the novel genes predicted by both groups are largely nonoverlapping. To validate the existence of the transcript predictions, we used RNA expression profiling and a bank of 13 diverse human tissues. The commercial high-density oligonucleo-The recent description of the human genome and the tide arrays used are based on Expressed Sequence subsequent annotation of putative novel genes has Tags (ESTs) represented in Unigene (release 95). ushered in a new era in biology. One of the revelations of BLASTN was used to assign the transcript predictions the human genome project was the remarkably consistent to a Unigene cluster, and the RNA expression pattern prediction that the genome harbors around 30,000 genes. was determined for the 8,000 known and 5,000 novel This observation was based on independent analyses predicted genes with a corresponding Unigene cluster done by a public genome consortium (29,691 transcripts, on the arrays (see legend to Figure 2 for details). Using Ensembl v0.8) (Lander et al., 2001), by work done at Celera these methods, we found evidence of expression for Genomics (39,114 transcripts) (Venter et al., 2001), and by more than 80% of the known genes in at least one of Green and colleagues using expressed sequence tag the tissue samples analyzed (Figure 2A). Similarly, more (EST) clustering incorporating quality scores (35,000 than 80% of the novel predicted transcripts were genes) (Ewing and Green, 2000). This conclusion was detected as expressed in at least one of the 13 human surprising for two reasons. First, less complex organisms tissues. Hierarchical clustering and visualization of like Arabidopsis (25,000) and C. elegans (19,000) have these expression data revealed a similar fraction of approximately the same number of genes (C. elegans tissue-restricted transcripts for both known and novel Sequencing Consortium, 1998; Arabidopsis Genome genes (Figure 2B). These data support the view that the Initiative, 2000). Second, earlier estimates of gene number novel transcripts predicted by both groups encode bona based on EST clustering and detailed chromosomal fide differentially expressed mRNAs. Since many of analysis were much higher, ranging from 45,000 to 140,000 these verified transcripts were contained in only one of the two predicted transcriptomes, we conclude that the Scott, 1999). While the Celera and Ensembl annotation computational methods used for gene prediction by efforts predicted approximately the same number of either group are inadequate and that the respective genes, a direct comparison of the predicted transcript sets transcriptomes are individually …

Transcript annotation in FANTOM3: mouse gene catalog based on physical cDNAs.

The international FANTOM consortium aims to produce a comprehensive picture of the mammalian transcriptome, based upon an extensive cDNA collection and functional annotation of full-length enriched cDNAs. The previous dataset, FANTOM2, comprised 60,770 full-length enriched cDNAs. Functional annotation revealed that this cDNA dataset contained only about half of the estimated number of mouse protein-coding genes, indicating that a number of cDNAs still remained to be collected and identified. To pursue the complete gene catalog that covers all predicted mouse genes, cloning and sequencing of full-length enriched cDNAs has been continued since FANTOM2. In FANTOM3, 42,031 newly isolated cDNAs were subjected to functional annotation, and the annotation of 4,347 FANTOM2 cDNAs was updated. To accomplish accurate functional annotation, we improved our automated annotation pipeline by introducing new coding sequence prediction programs and developed a Web-based annotation interface for simplifying the annotation procedures to reduce manual annotation errors. Automated coding sequence and function prediction was followed with manual curation and review by expert curators. A total of 102,801 full-length enriched mouse cDNAs were annotated. Out of 102,801 transcripts, 56,722 were functionally annotated as protein coding (including partial or truncated transcripts), providing to our knowledge the greatest current coverage of the mouse proteome by full-length cDNAs. The total number of distinct non-protein-coding transcripts increased to 34,030. The FANTOM3 annotation system, consisting of automated computational prediction, manual curation, and final expert curation, facilitated the comprehensive characterization of the mouse transcriptome, and could be applied to the transcriptomes of other species.

A Gene Wiki for Community Annotation of Gene Function

This manuscript describes the creation of comprehensive gene wiki, seeded with data from public domain sources, which will enable and encourage community annotation of gene function.