Sybil: Web-based software for comparative genomics

Sybil is a web-based software package for comparative genomics. It was developed by the Bioinformatics department at The Institute for Genomic Research (TIGR) and development continues at the J. Craig Venter Institute and Institute for Genome Sciences (IGS) at the University of Maryland Baltimore's School of Medicine. The project has been funded by a variety of sources including: The National Institute of Allergy and Infectious Diseases (NIAID) as part of the NIAID Bioinformatics Resource Centers (BRC) for Biodefense and Emerging or Re-Emerging Infectious Diseases and PATH an organization supporting research in Streptococcus and Neisseria vaccine development.

While still under development, several of the Sybil tools have been used internally at TIGR, the JCVI and IGS by research projects that involve comparative genome analysis. See the SCREENSHOTS section below for examples of the types of analyses and displays that the system currently supports, or follow the link to the Demos page to interact with a live (albeit scaled-down) version of the system, running on a demonstration database that contains genome sequences, annotation, and comparative data from ten Streptococcus pneumoniae genomes. If you have any questions or suggestions, please contact us using the sybil-info mailing list.

Screenshots

The following screenshots illustrate some of the displays and web pages available to a user who wishes to interactively explore one or more comparative datasets using the Sybil web interface. Note that all of the examples on this page are accompanied by precomputed static screenshots; for those displays that have been released in the downloadable source code, links ("live demos of this display") are provided to one or more example pages on the live Sybil demonstration web site.

Interactive graphical displays

The following displays make extensive use of simple 2D graphics to convey information from a single chado comparative database. Almost all of these graphical displays are clickable, allowing the user to navigate between them using only the mouse. Some of the displays provide links to download the graphical elements in PDF or SVG format.

Whole Genome View

live demos of this display: S. pneumo site
full-size image: whole_genome_view_cropped.png

This display provides a simple overview of the selected genome annotations along with the location or density graph of any additional feature types loaded such as repeats, SNPs, PMARKs etc. This view supports display of %GC as well as GC-skew plots. Clicking on a region of the genome will generate a popup box that contains pertinent information and links to other displays. This view is still under active development.

Synteny Gradient View

live demos of this display: S. pneumo site
full-size image: pneumo_gradient.png

This display shows synteny between the selected genomes based on a chosen reference. The genes in the reference are colored from yellow to blue from left to right. The matching genes (based on ortholog clustering) in the query genomes are positioned overtop of the reference gene they hit and are colored based on where they are located in their respective genomes. This display can show where a set of genomes share conserved gene order (synteny) as well as where they diverge. It can highlight rearrangement events strain/species specific regions. Clicking on a gene in the reference provides a popup that contains pertinent information and links to other displays. This view is still under active development.

Protein cluster display

live demos of this display: S. pneumo site
full-size image: protein_cluster.jpg

This page provides details about a particular gene cluster (orthologous or paralogous). At the top of the page are some statistics about the cluster including the average percent identity and coverage of the BLAST hits that went in to generating it. Below the statistics is a list of the genes/proteins that have been grouped together by the cluster The next item on the page is a graphical display of the genes whose protein products have been clustered; these "clustered" genes appear in the center of the image and are connected by regions shaded in red/pink. Nearby genes on each of the genomic sequences are also shown, and those that were also clustered together (by the same clustering analysis) are connected by regions that are shaded in grey.

Clicking on any gene displays a popup with pertinent information and links to other sybil pages related to that gene. Clicking on any of the red or grey shaded regions will display a popup showing the members of that protein cluster in addition to a link to that protein cluster page. The image can be output as an SVG.

Below the context image are pictorial representations of the best BLAST hits for all the cluster members selected at the top of the page. This image shows the alignments of each of the best hits and provides links to their respective protein report pages.

At the bottom of this page is a multiple sequence alignment that can be rerun to reflect the cluster members selected in the list. This alignment can be requested for either the polypeptide or transcript DNA.

Protein/gene report

live demos of this display: S. pneumo site
full-size image: protein_report.png

This page displays information pertaining to a single gene and its protein product, including:

external database references and any alternate identifiers stored in the database
a graphical display of the gene's genomic localization(s)
a list of the protein/gene clusters to which the gene belongs
a graphical representation of aligned best blast hits
the gene's protein sequence, locus name and primary accession number

Clicking on one of the other genes in the genomic context display will move to the protein/gene report page for that gene.

Genomic region comparison

live demos of this display: S. pneumo site
full-size image: genomic_comp.png

This view resembles the protein cluster report page above but is more flexible. On this page you chose a reference sequence region an then several target genomes. The regions of target genome that share above a certain number of clusters with the reference will be drawn above the reference and grey links will be drawn between matching genes. Like the protein cluster page, the genes and cluster links are clickable. Clicking one of these features generates a popup with information about the feature clicked as well as links to other Sybil page.