FlyBase:Overview

Expresson Data

Separate Gene Expression Reports have been retired from FlyBase in favor of integrating these data into the Gene Report in a dedicated ‘Expression Data’ section (Fig. 2). Expression data may derive from either ‘low-’ or ‘high-throughput’ studies.

Low-throughput expression data

Traditional expression assays, such as in situ hybridization or immunolocalization to embryos or tissues, and stage- or tissue-specific Northern blots, are presented in the first three subsections of the ‘Expression Data’ section: ‘Transcript Expression’, ‘Polypeptide Expression’, and ‘Expression Deduced from Reporters’. These are data from the published literature, captured in a highly controlled format using the FlyBase anatomy and developmental stage CVs. For nuanced aspects of an expression pattern additional free-text descriptions are provided. Embryonic transcript expression data include data from the BDGP (18), which comprises in situ hybridizations using cDNA probes for over 7,000 genes, plus descriptions conforming to the FlyBase anatomy CV. The actual in situ images can be viewed by following the ‘BDGP expression data’ link in the subsection ‘External Data and Images’ (see section 5.3).

Querying low-throughput expression data: The use of hierarchical CVs allows expression data captured at a very detailed level to be queried using more general terms. The QuickSearch ‘Expression’ tab provides a dedicated interface for this type of query. It also allows combinatorial queries, typically to specify both stage and tissue. The initial hit list returned by this query is of endogenous genes for which the expression pattern is observed; reporter constructs or insertions can be retrieved by selecting one of the alternative result options at the top of the initial hit list.

High-throughput expression data

Within the ‘Expression Data’ section of the Gene Report, the ‘High-Throughput Expression Data’ subsection includes expression plots of high-throughput mRNA RNA-Seq data from modENCODE (19) and mRNA microarray data from FlyAtlas (20). FlyBase has produced quantitative views of these data in different stages, tissues, and cell culture types presented as bar graphs (Fig. 3). For the RNA-Seq data, RPKM counts (21) (reads per kilobase per million reads) have been calculated, averaged over the exonic extents of the gene. A further subsection, ‘Expression Clusters’, provides links to datasets consisting of genes possessing similar mRNA RNA-Seq expression dynamics, as determined by modENCODE (22,19).

Querying high-throughput expression data: Several new FlyBase tools that use the calculated RPKM data have been developed. The versatile ‘RNA-Seq Profile’ tool allows retrieval of genes with user-defined RNA-Seq expression patterns and levels (values are binned); this tool can be accessed from a pictograph on the homepage, from the Tools menu of the NavBar, or from the QuickSearch ‘Expression’ tab. A section of the QuickSearch ‘Expression’ tab also provides options to ‘Search for similarly expressed genes’, using the modENCODE RNA-Seq datasets. The ‘RNA-Seq by Region’ tool returns the average RPKM over a specified genomic region and also offers the option of a gene-specific query that returns an exon-by-exon RPKM count; this tool can be accessed from the Tools menu or from the ‘High-Throughput Expression Data’ subsection of the Gene Report.

External link-outs

The final ‘Expression Data’ subsection on the Gene Report is ‘External Data and Images’, with gene-specific links to other databases that include expression data for Drosophila. There are links to the original BDGP in situ data, as well as FlyExpress (23) analyses that use the BDGP data and allow a ‘Find Similar Patterns’ option. The FlyAtlas link provides the underlying microarray data used for the bar graphs described above. For the FlyExpress and SliceSeq (24) databases, sample images are shown.

Interactions

Both genetic and physical interaction data are presented in FlyBase. The former are primarily recorded in Allele Reports, while the latter are given in dedicated Physical Interaction Reports; both are summarized in the ‘Interactions and Pathways’ section of the relevant Gene Report page. Both types of interaction can be viewed either as a graphical ‘network diagram’ provided by esyN (25) or within the FlyBase Interactions Browser tool that includes additional viewing and configuration options (4). The ‘External Data’ subsection of the ‘Interactions and Pathways’ part of the Gene Report provides link-outs to relevant pages at third-party interaction databases, including BioGRID (26), DroID (27) and InterologFinder (28).

Genetic interactions

Genetic interaction data are recorded at the allele level using phenotypic class and anatomy CV terms (and optional qualifiers), similar to phenotype annotations but with the addition of terms such as ‘enhanced by’ or ‘suppressor of’ to indicate the nature of the interaction, together with the interacting allele. (Negative results, e.g., ‘not enhanced by’, are also captured.) In addition to enhancer/suppressor-type interactions, synthetic phenotypes that are present in a mutant combination but absent in single mutant conditions are also captured. All these interaction statements appear in the ‘Interactions’ section of the Allele Report, alongside free text clarifications where necessary. All genetic interaction statements in FlyBase are curated from the published literature.

The allele-level genetic interaction statements are used to compute a gene-level summary of these data, and this appears as a table within the ‘Summary of Genetic Interactions’ subsection in the Gene Report. This table shows the interacting genes, the nature of the interaction (limited to enhancer/suppressor-type interactions) and the reference(s) supporting the interaction. These gene-level interactions are used to power the esyN network diagram.

Querying genetic interactions: Both allele-level and gene-level genetic interaction data can be queried directly via the Interactions Browser tool. More specific and/or combinatorial searches may be conducted using QueryBuilder.

Physical interactions

The Physical Interaction Report displays pairwise physical interaction data for gene products, either protein-protein or RNA-protein. Each report includes the experimental assays used (e.g., co-immunoprecipitation, peptide mass fingerprinting), the role of each protein in an assay (e.g., bait or prey; whether a tagged or endogenous protein was used), the esyN network diagram, and a link to the Interactions Browser tool. With the goal of producing a set of high-confidence pairwise interactions, our current focus is on smaller-scale physical interaction data curated from the literature, which usually include multiple types of support for a described interaction. High-throughput interaction datasets are curated only when the authors take care to filter out false positives--criteria for curation may include: multiple negative control purifications, accounting for protein abundance in assessing the likelihood that a purified factor is a contaminant, a calculation of the confidence level, and an explicit cut-off to separate high confidence and lower confidence interactions. Examples are the DPiM dataset (29), the Hippo Pathway Interactome (30) and the ECIA extracellular interactome (31).

Within the ‘Summary of Physical Interactions’ subsection of the Gene Report, all pair-wise physical interactions involving that gene product are presented in tabulated form, with assays used, attributed publications, and links to the corresponding Physical Interaction Report.

Querying physical interactions: The ‘Simple’ or the ‘Data Class (physical interactions)’ tabs of QuickSearch can be used to find interactions involving a given gene or to search for assay terms present in the Physical Interaction Report.

Genomic Data

Genomic data in FlyBase comprise gene model annotations (i.e., the exon-intron structure and transcription and translation start/termination sites of genes) and any other sequence-based features that can be mapped to specific genomic coordinates, whether endogenous (e.g., regulatory regions, origins of replication) or describing a lesion/reagent (e.g., insertion sites, RNAi amplicons). All these data are viewable through the FlyBase implementation of GBrowse (32) (Fig. 4). Many are also associated with discrete Sequence Feature Reports and are searchable through the FeatureMapper tool (11) (Fig. 5). Note that FlyBase currently uses release 6 of the sequenced D. melanogaster genome (33) – the Coordinates Converter tool, accessible from the Tools menu of the NavBar, can be used to convert data from release 3, 4 or 5 coordinates (32).

Gene model annotations in FlyBase

For D. melanogaster, FlyBase has produced manually annotated gene models for over a decade (34). Since 2010, RNA-Seq data (35,19) and new transcription start site data (33,36) have supported many major changes in the gene model annotations for this species. This prompted a comprehensive review of all existing gene models and the annotation of several thousand new genes, primarily long non-coding RNA genes (34). Transcript and protein data are tabulated in the ‘Gene Model and Products’ section of the Gene Report (Fig. 2), with links therein to more detailed reports. D. melanogaster gene models continue to be updated regularly based on new high-throughput and literature-based data. An updated gene model set is submitted to GenBank approximately once a year and serves as the NCBI RefSeq set for this species.

For eight of the other sequenced Drosophila species (D. ananassae, D. erecta, D. pseudoobscura pseudoobscura, D. simulans, D. yakuba, D. mojavensis, D. virilis and D. willistoni), the long-standing CAF1-generated gene model annotations (37) have been replaced recently by sets generated by NCBI as part of their GNOMON annotation pipeline (38). Gene model annotations for three other species (D. grimshawi, D. persimilis and D. sechellia) have not been updated owing to poor genome assembly quality or to lack of RNA-Seq data, which provides the primary basis for robust annotation by the GNOMON pipeline.

Querying gene model data: Gene models can be searched directly in GBrowse (section 7.2) or via the CytoSearch or FeatureMapper tools (both accessible via the Tools menu of the NavBar). The FlyBase BLAST tool allows sequence-based queries against annotated transcripts or proteins from the twelve Drosophila species mentioned above. For D. melanogaster, Sequence Ontology terms and controlled comments have been used extensively to describe gene models and transcripts (34,39). These enable queries for exceptional cases, such as all genes with dicistronic transcripts or all transcripts annotated with non-canonical translation start sites, using QueryBuilder or the ‘Simple’ tab of QuickSearch. ‘Transcripts’ and ‘Polypeptides’ are data class options in QuickSearch and QueryBuilder, thus allowing class-specific and field-specific queries of gene products, respectively.

GBrowse

GBrowse, a genome annotation viewer that is part of the Generic Model Organism Database (GMOD) tool suite (40), is used by FlyBase to show gene models and supporting data, such as cDNAs, ESTs, RNA-Seq data, transcription start sites, gene predictions, and aligned proteins (32) (Fig. 4). In addition, this versatile tool allows representation of many other types of sequenced-based data and reagents--essentially anything that maps to the genome can be represented on GBrowse. By using the ‘Select Tracks’ option, the user can choose to view mapped genetic variants such as mutational lesions, transgenic insertions, aberration extents, and aberration breakpoints; regions carried on transgenic constructs such as rescue fragments and RNAi reagents; microarray oligonucleotides and RNAi amplicons; or high-throughput mapping of transcription-factor binding sites, insulator elements, and RNA-editing sites. When zoomed in to a range of 100–200bp, the tracks indicating forward/reverse translation and ‘DNA/GC content’ switch to the nucleotide or protein sequence. The current version of this tool is GBrowse2 (41), which allows rapid customization options: for example, a selected track can be moved by simply dragging the track title bar vertically, and tracks can be closed, opened or removed using the icons in the track title bar. Moreover, navigation within a genomic region has been facilitated by limited smooth-track panning (side-to-side sliding) and by a function that allows the user to lasso a smaller region and zoom in.

RNA-Seq expression data (19,42) are particularly informative when viewed in GBrowse. In 2010, FlyBase debuted a new topographical presentation of these data for GBrowse that allows visual assessment of many RNA-Seq tracks at once (Fig. 4). By clicking on the wrench/spanner icon in the track title, the presentation can be changed from log2 to linear, and from tilted to vertical; specific tracks corresponding to different tissues and/or development stages can be shown or hidden. RNA-Seq exon junction data (19,35), presented in a separate track on GBrowse, are extremely useful for judging alternative splicing and isoform-specific expression.

GBrowse supports a number of download options, accessible from the drop-down menu on the upper right of the page, including a FASTA file of the sequence shown and an HTML table view or a GFF file of all the mapped genes and features selected. The sequence of a lassoed genomic region can also be viewed and downloaded.

GBrowse can be accessed from one of the pictograph buttons at the top of the homepage or via the Tools menu on the NavBar. In addition, there is a link to the appropriate genomic region in GBrowse on the reports for every localized gene and mapped sequence feature. A genomic BLAST hit obtained using the FlyBase BLAST tool also includes a link to the relevant region in GBrowse.

Sequence features and other genomic data tracks

‘Sequence features’ are defined as regions of DNA/RNA that can be mapped to the genome sequence and to which a discrete function can usually be ascribed. They include endogenous regions such as enhancers, insulators, transcription factor binding sites, transcription start sites and origins of replication, as well as experimental reagents that map to the genome, such as RNAi reagents and putative enhancer element constructs. Sequence features appear within discrete tracks on GBrowse and are associated with dedicated report pages. (Note that certain GBrowse tracks, including point mutations, transgenic insertions and aberration extents, are not classed as ‘sequence features’ and are instead associated with specific Allele, Insertion or Aberration Reports.) Most sequence features currently in FlyBase were generated in response to the modENCODE project (22) and similar large-scale experiments (43,44).

The Sequence Feature Report is flexible, in order to accommodate many different types of genome-associated data. The typical report includes a link to the Large Dataset Metadata Report (see section 9.1) to which it belongs, the sequence itself and its genomic location, a genome snapshot showing the alignment of that feature alongside other sequence features included within that region, and links to any relevant external websites/databases. Clicking on the ‘GBrowse’ link near the top of the page goes to a full genome view of the respective region in GBrowse.

Querying sequence features and other genomic data tracks: Limited querying can be performed within GBrowse itself by specifying a ‘Landmark or Region’ and selecting particular tracks for display. A better approach is to use FeatureMapper (Fig. 5), which provides an intuitive interface for retrieval of specified genome features in one or more genomic regions, with results presented in a convenient table that includes an option to export to a hit list where possible. The CytoSearch tool allows retrieval of genes, aberrations, and transgenic insertions mapped to the genomic sequence. Sequence features are also included as a specific option in the ‘Data Class’ tab of QuickSearch.

Reagents

There are several ways to find reagents associated with a specific gene or genomic region. The ‘Stocks and Reagents’ section of the Gene Report is a good place to start. Here, subsections list publicly available fly stocks, genomic and cDNA clones, cell-based RNAi reagents and antibodies described in the published literature (Fig. 2). Other reagents are best found by searching a genomic region of interest using GBrowse, FeatureMapper or CytoSearch. For example, the GMR (45) and VDRC (44) putative enhancer collections are not associated with specific genes, while some classes of transgenic insertions are not listed in the Gene Report. Moreover, a visual representation of the location of a sequence-based reagent relative to the gene of interest is often informative when planning experiments.

Stocks

Stock Reports display the stock list genotype and the source collection, together with the stock number hyperlinked to the specific record at the appropriate stock center to facilitate ordering. There are links to Stock Reports from other appropriate reports (primarily alleles, aberrations, transgenic constructs and insertions) throughout FlyBase. The Bloomington Drosophila Stock Center is the most widely represented source, though many others are included--a complete list can be found in the ‘Links’ menu on the NavBar.

Querying stocks: Stocks can be searched specifically by selecting ‘stocks’ in the ‘Data Class’ tab of QuickSearch.

Strains

FlyBase Strain Reports contain data about wild type strains such as ‘Oregon-R’, significant mutant strains such as ‘iso-1’ (the D. melanogaster strain sequenced by the BDGP (33)), as well as the 200 or so inbred lines generated by the Drosophila Genetics Reference Panel (46). The reports include information on the origin and history of the strain alongside any known genetic or phenotypic components (e.g., the ‘iso-1’ strain harbors several mutations). Where relevant, links are also provided to Large Dataset Metadata Reports (section 9.1) that describe strain collections, and to Stock Reports to facilitate ordering. (Note that stocks are instances of strains in theory, but they are effectively distinct in time and place and may have characteristics that differ from the strains from which they descended.)

Querying strains: Strains can be searched using the QuickSearch ‘Simple’ tab.

Cell lines

Cell Line Reports display data obtained from the Drosophila Genomics Resource Center (DGRC) on cell lines, such as ‘Kc167’ or ‘S2R+’. The reports include the source and development stage of each line, its sex and karyotype (where known), and any parental or descendent lines. A link back to the DGRC is also provided for additional data and ordering information.

Querying cell lines: Cell lines can be searched specifically by selecting ‘cell lines’ in the ‘Data Class’ tab of QuickSearch.

cDNAs

cDNAs are shown in GBrowse and appear in the ‘Stocks and Reagents’ section of the Gene Report of the aligned gene(s). Links from GBrowse go to the GenBank report; links from the Gene Report go to the FlyBase Clone Report. The Clone Report includes the sequence, links to GenBank, and fields for ‘Known Problems’ and ‘FlyBase assessment’. Examples of known problems are clones that are chimeric or that contain genomic DNA or transposon sequences. The FlyBase assessment field displays a note if the clone has been replaced, for example “Caution: This cDNA clone replaced by FI01005”. There is also a link to the DGRC where clones are available from that resource.

Querying cDNAs: cDNA clones can be searched specifically by selecting ‘clones’ in the ‘Data Class’ tab of QuickSearch. FeatureMapper should be used to find cDNAs associated with a specific gene or genomic region.

Integrated Reports

As the amount of Drosophila data and resources increase in FlyBase, it has become both necessary and useful to organize and integrate related data into discrete sets or collections. This has multiple benefits, including the ability to associate metadata across a range of related entities, and to present related data to users in new ways that aid comprehension. To date, FlyBase has developed three types of such integrated reports.

Large Dataset Metadata

Large Dataset Metadata Reports, previously named Library/Collection Reports, provide information on large datasets and reagent collections that apply to the set as a whole. Examples of datasets are the protein interaction network defined by the Drosophila Protein interaction Mapping (DPiM) project (29), the set of RAMPAGE transcription start sites (36), and datasets generated by the modENCODE project (22). Examples of collections are the set of dsRNA amplicons used for RNAi-knockdown assays in cell culture by the Drosophila RNAi Screening Center (49), the set of defined X-chromosome duplications made by the Bloomington Stock Center (50), and several large construct and insertion collections. Metadata describing cDNA libraries are also captured in this format. The Large Dataset Metadata Report includes the type of dataset or collection, a brief description of the set, a summary of the experimental details, and a link to download all the associated features. Links to external data repositories and reagent sources are provided where relevant. The ‘Description’ field of the dataset report is propagated to each member report; reciprocal links are provided.

Querying large dataset metadata: The ‘Simple’ or the ‘Data Class (large dataset metadata)’ tabs of QuickSearch can be used to find datasets and collections of interest.

Gene Groups

Gene Group Reports have been introduced to allow easy access to, and analysis of, related sets of D. melanogaster genes and their associated data (47). Examples of gene groups include members of a gene family (Actins, Wnts…), subunits of a protein complex (proteasome, ribosome…), or other functional groupings (protein kinases, Ubiquitin E3 ligases…). All gene groups in FlyBase are based on published literature and the basis for the membership of each group is clearly attributed. The main feature of these reports is a ‘Members’ table that lists the genes comprising the group, arranged into a series of subgroups where appropriate. Buttons are provided to facilitate the downloading of associated data (phenotypes, expression data, protein interactions etc.) using Batch Download (section 10.2), or to further refine or analyze the gene set by exporting it to a standard hit list. Also shown are links to equivalent gene groups for other organisms, including nematodes (WormBase (48)) and humans (HGNC (8)). To aid navigation, the ‘Families, Domains and Molecular Function’ section of the Gene Report contains a link to any associated gene group(s) (Fig. 2).

Querying gene groups: Gene groups can be retrieved by entering the symbol/name of a group or any member gene in the ‘Gene Groups’ tab of QuickSearch. This tab also includes a link to a browsable list of all current gene groups in FlyBase.

Human Disease Models

Human Disease Model Reports provide a less specialized entry point into FlyBase for researchers interested in Drosophila models of human disease (17). Data from numerous outside sources, including OMIM, and from recent reviews are presented in a general ‘Disease Summary’ section, followed by information on orthology between a human gene implicated in the disease and the related Drosophila gene(s). For many diseases, multiple causative genes have been implicated; OMIM describes these as different disease subtypes and groups them into ‘phenotypic series’. In the Human Disease Model Report, such a phenotypic series of subtypes is presented in a table titled ‘Related Diseases’, which includes links to other relevant Human Disease Model Reports and provides a quick view of which disease subtypes have been modeled in flies.

The major portion of the disease report is devoted to ‘Experimental Findings’ in Drosophila, focusing on disease-related implications and results. Descriptions of specific experiments are meant to be generally accessible, with links to Allele Reports with more detailed information. Results may include data using both fly genes and human genes introduced into flies. The ‘Experimental Findings’ section initiates with a FlyBase-authored summary that presents a concise review, including phenotypes, interactions, and suitability of the model for drug assays; in addition, new findings and emerging mechanistic themes are highlighted. At the end of this section, a link to the FlyBase Disease Wiki is provided; comments and contributions from users are encouraged, especially those with expertise in the specific disease model. The last sections of the report draw relevant data from other sections of FlyBase, including physical interaction data for the orthologous Drosophila gene(s), a table of genetic reagents and stocks useful for investigations of human disease, and a table of Disease Ontology-based annotations of alleles used for that disease model (see section 4.3).

There are links to relevant Human Disease Model Reports in the ‘Human Disease Model Data’ section of Gene Reports (Fig. 2). Note that many such links are found in FlyBase Gene Reports for human genes (e.g., Hsap\SNCA and Hsap\TARDBP).

Querying Human Disease Model Reports: These reports can be found by using the ‘Human Disease’ or ‘Simple’ tabs of QuickSearch, or by searching the Disease Ontology within the Vocabularies tool.

Bulk Data Analysis and Downloads

Users increasingly want to be able to process data in bulk. They may have generated a hit list of genes (or any other data class) within FlyBase, or have a list of IDs from elsewhere to upload, and wish to analyze/refine this list or obtain associated data. Alternatively, users may wish to directly obtain bulk data files corresponding to a particular data type for processing off-line.

Uploading and analyzing data

A list of IDs (e.g., gene symbols or CG numbers, allele or insertion symbols, FlyBase identifiers) can be pasted or uploaded into the Upload/Convert IDs tool (Fig. 6; accessed via the Tools menu on the NavBar). This tool will then validate the list, updating any obsolete IDs to the current version where possible, and generate a ‘Conversion report’ clearly indicating if any of the submitted IDs failed verification. The user can choose to correct these cases, or ignore them before proceeding to convert the list into a standard FlyBase hit list (see section 2.2). This list can then be further analyzed/refined before being exported or downloaded as required.

Downloading data

Batch Download is a powerful tool for generating customized output files in various formats for most data types in FlyBase (11). Users may arrive at Batch Download via a hit list (as described above), by navigating to it from the Tools menu of the NavBar, or by clicking on its pictograph on the homepage. If the first, then the input list will be pre-filled (Fig. 6); otherwise the user can paste in or upload a list of symbols or IDs directly. Depending on the nature of the input and the desired outcome, the output format can then be specified as ‘FASTA Sequence’ (with the option to further specify introns, UTRs, CDS etc.), ‘Database Format’ (XML), or as ‘Field Data’ (with output options of an HTML table, a tab-separated value (tsv) file, or in the same format as the precomputed files described below). If the ‘Field Data’ option is selected, the user can then specify any combination of data fields (appropriate to the given data class) from a page styled in the same format as a standard FlyBase report page.

Bulk files of FlyBase data can be downloaded using our FTP site (ftp://ftp.flybase.org/releases/) or the ‘Downloads’ menu of the NavBar on the website (see the ‘Overview’ page under the Downloads menu for more details). ‘Precomputed files’ contain particular slices of FlyBase data that users or collaborators have requested over the years or are otherwise difficult to obtain in bulk (Table 3). Notable recent additions include D. melanogaster unique protein isoforms, RPKM gene expression values, gene groups, and physical interactions. Also included are several useful correspondence tables and the ontology files used in FlyBase (Table 3). In addition, Chado XML (database format) files are provided for all FlyBase data classes and comprehensive sets of FASTA, GFF and GTF files are available for the twelve originally sequenced and annotated Drosophila species (see section 7.1). The FASTA files comprise many different cuts of genomic data, including annotation categories such as small RNA classes and pseudogenes, components of gene model annotations such as exons, introns, UTRs and predicted translations, as well as other genome features such as transposons and intergenic sequences. As described above, Batch Download can also be used to obtain specified subsets of data in precomputed file, Chado XML or FASTA format by selecting the appropriate output options.

Most bulk files are regenerated for every release of FlyBase. Those corresponding to the current or previous (archived) versions of FlyBase are found under the appropriate submenus/subfolders on the web/FTP site. The release version used for a particular file is indicated in the file name and in the header lines of the file itself.

The FlyBase Community

FlyBase engages with our user community through multiple approaches. The primary method for users to get in touch with FlyBase about any matter remains our ‘Contact FlyBase’ page, accessible via the ‘Help’ menu on the NavBar or the link in the footer of any FlyBase page. All other community resources are grouped under the ‘Community’ menu of the NavBar and/or are found on the homepage.

If a user wants to specifically alert us to a Drosophila publication or data therein to be added to FlyBase, then the ‘Fast Track Your Paper’ (FTYP) tool should be used (51). This tool allows the user to indicate the key genes studied and flag data types present in a paper. The resulting gene-to-publication links are submitted directly to the FlyBase database while the data type information is used to prioritize the paper for more detailed curation. We actively solicit FTYP submissions using our ‘EmailAuthor’ pipeline, whereby the corresponding author of a Drosophila publication is automatically sent an email that includes a link to a personalized FTYP form (51). Approximately 50% of authors respond to this request, thereby reducing by half the amount of manual triaging to be done by FlyBase curators.

Our recently launched ‘FlyBase Community Advisory Group’ (FCAG) is a worldwide group of over 500 volunteers (lab heads, postdocs, students, technicians) who use FlyBase for a range of purposes. We contact this group up to six times per year with a survey on a variety of subjects to get feedback about how data collection, presentation and searching on FlyBase can be improved. By consulting this relatively large, diverse group of researchers, we hope to implement changes to FlyBase that are helpful for the greatest number of people.

Users may also help improve FlyBase by contributing to the Human Disease Wiki (described in section 9.3) or the FlyGene Wiki. There is a link to the latter at the top and within the ‘Summaries’ section of each Gene Report. This is pre-seeded with the automatically-generated FlyBase summary and users are encouraged to modify or add to this text to build up a more complete and readable summary of each gene’s main features and functions.

The FlyBase Forum is a Google™ Group that provides an alternative, more open platform for users to interact both with FlyBase and with each other. The forum has two areas: one for general questions and discussions about FlyBase and Drosophila protocols etc., and the other for relevant job postings.

Users are made aware of new or changed features in FlyBase through any of several means. First, there are the ‘News’ and ‘Commentary’ sections of the FlyBase homepage (Fig. 1). Second, users can sign up to receive an occasional Newsletter via email by clicking the link on the homepage. The Newsletter contains release announcements, significant website updates, and other important Drosophila community news. Third, to obtain more frequent updates, users can follow FlyBase on Twitter™ by clicking on the icon in the footer of any FlyBase page. Fourth, users can choose to subscribe to any FlyBase record (a specific gene, transgenic construct, reference etc.) and receive automatic updates through a feed reader by clicking the icon in the ‘Recent Updates’ section of any report page. Finally, users have the opportunity to see and hear about FlyBase updates in person at the Annual Drosophila Research Conference in the USA and the biennial European Drosophila Research Conference, where FlyBase representatives give presentations and are available to answer questions. Previous conference presentations and pamphlets can be obtained via the ‘FlyBase Guides’ link under the ‘Help’ menu in the NavBar.