FusoPortal by The Slade Lab

What is FusoPortal?

Fusoportal is an online database of Fusobacterium genomes sequenced by The Slade Lab at Virginia Tech using an Oxford Nanopore MinION and Illumina reads, and assembled using the software Unicycler developed by Ryan Wick in the group of Kat Holt. All genomes are annotated using Prodigal, and proteins are annotated using HMMER, InterPro, and Gene Ontology (GO) terms via Blast2GO. All data is made available as clickable and downloadable links to files found in the whole genome map or pages for individual genes. Why not take a look at the first full assembly of F. nucleatum 23726, a genetically tractable bacterium that is biomedically relevant in multiple diseases including colorectal cancer.

In addition, all raw data is available under NCBI BioProject PRJNA433545

What genomes are available?

The first genome we sequenced and provide here is Fusobacterium nucleatum subsp. nucleatum (strain ATCC 23726 / VPI 4351). We chose this genome because it is genetically tractable, but we found several limitations of previous genomes on the internet that were in 67 contigs, and contained breaks in several genes we were studying. We will be expanding FusoPortal to include more genomes in the near future. Access all available genomes here.

We have built a custom BLAST server to easily search these genomes.

Want to screen one of your proteins to find out if there is a copy in a Fusobacterium genomes? We have implemented a custom FusoPortal BLAST server that will allow you to search one, two, or all of our genomes simultaneously. Results can be cross referenced against our whole genome maps, and alignments and fasta files for individual genes are available in downloadable formats.

What need does FusoPortal fill?

While online databases including KEGG and Uniprot are very useful for initial discovery of genes, we found that they lack accuracy in genome annotation; especially in large proteins which make up a large portion of Fusobacterium virulence factors. These shortcomings are due to a combination of incompletely assembled genomes, and open reading frame prediction software that is not optimal for bacteria. We aimed to improve upon this by creating completely assembled genomes, followed by using genome annotation software programs that is specific for bacterial genomes. We hope you enjoy and find it useful! If you would like to compare our accuracy and annotation, please read our papers:

Sanders et al. and Todd et al.

FusoPortal genome assembly and annotation work flow

Here is a downloadable PDF of the tree above that contains links to each genome and sequence reads in the databases in which they were deposited. This tree was adapated from Manson McGuire et al, mBio 2014, from the laboratory of Ashlee Earl.

How are these genomes assembled and annotated?

We used some amazing open source software by top developers. This pipleline is highly similar to that described by Wick et al. This robust system, which works well with multiplexing in both the MinION and Illumina runs, allows genomes to be sequenced, assembled, and annotated in house in a cost effective manner. We estimate that even with having the Illumina library prepped and sequenced at our core facility at Virginia Tech, the cost of each genome is <$500 US.

Want to find a specific protein or homolog? Use our custom BLAST server.

To make all of our data searchable, we created a custom BLAST database in which you can search one, two or all of our genomes at once using either DNA or amino acid sequences. BLAST hits are downloadable as .fasta files, and proteins hits and their e-Value and bit scores can be downloaded in full tabular or XML reports. We have to give serious credit to The Wurm Lab for creating SequenceServer, the framework from which this database was created.