Research
- ARDB-Antibiotic Resistance Genes Database.
(Now it's available at http://ardb.cbcb.umd.edu/)
Efforts on identifying new antibiotics were once a top research and development priority among pharmaceutical
companies. However, the treatment of infections is increasingly compromised by the ability of bacteria to
develop resistance to antibiotics through mutation or acquisition of resistance genes. Treatment failure can
have serious consequences in patients with infections of some pathogens, such as methicillin-resistant
Staphyococcus aureus (MRSA). Antibiotic resistance genes are also potentially
be used by bio-terrorists in genetically modified organisms. In order to facilitate identification and
characterization of antibiotic resistance genes, we have created a manually curated database (ARDB)
unifying most of the publicly available resistance genes and related information. Resistance genes are further
categorized to resistance types by their resistance profiles and sequence similarity. Each gene or type is annotated
with rich information, including resistance profile, resistance mechanism, resistance requirement, epidemiology,
GO term, COG and CDD. ARDB allows the user to browse and search antibiotic resistance information from a
view of gene, type, organism and antibiotic. Regular BLAST and RPS-BLAST tools would help the user to
identify and annotate new potential resistance genes. ARDB
can help user to identify mutational resistance for 12 antibiotic target genes. Currently, ARDB
contains resistance information for 13293 genes, 377 types, 257 antibiotics, 933 species and 124 genera.
- Antibiotic Resistance Genes Annotation and Resistance Profiles Comparison
- Single Gene Annotation. Three tools based on BLAST are available for single gene annotation
to meet the user's taste.
(1) ARDB
protein database BLAST server
(2)
ARDB nucleotide database BLAST server
(3)
ARDB Position Specific Scoring Matrix (PSSM) database RPS-BLAST server
- Multiple Genes Annotation
(Genome Annotation) and Comparison
This tool allows user to annotate multiple query
genes at once. Besides, if the query genes are from a newly sequence genome, then you can
compare the annotated resistance profile with other already sequenced pre-annotated genomes
(632 genomes).
- Antibiotic Resistance Profiles Comparison
of Pre-Annotated Genomes (632 Genomes)
This tool allows the user to compare the
resistance profiles of already sequenced genomes.
- Antibiotic Resistance Prediction from
Gene Mutation (initial version)
This initial version tool actually does prediction by memorization of training data. It maps
the query sequence against the reference sequence to find polymorphisms (mutations),
if these mutations have been observed in the training data to be conferring resistance,
then the query sequence is predicted to be resistant. Because the training data are very
heterogeneous and sparse, in order to generalize the prediction, a tool based on supervised
learning using artificial neural networks and unsupervised clustering is underway.
Publications
- ARDB-Antibiotic Resistance Genes Database. Bo Liu, Mihai Pop. Nucl. Acids Res. 2009 37: D443-D447
Presentations
- Antibiotic Resistance Genes Database and Evolution of Resistance Genes in Staphylococcus aureus genomes,
CBCB Works in Progress Seminars, University of Maryland, College Park, MD, 10/18/2008
- Systems approach for understanding cellular responses to gamma radiation in Halobacterium NRC-1
using Cytoscape, CBCB Seminar Series, University of Maryland, College Park, MD, 12/07/2006
Posters
- ARDB-Antibiotic Ressitance Genes Database,
The Biology of Genomes, Cold Spring Harbor Laboratory, NY, 05/08/2008
Courseworks
- Algorithms in Biosequence Analysis, Mihai Pop
- Computational Gene Finding and Genome Assembly, Steven Salzberg
- Biostatistics, Frank Siewerdt
- Bioinformatic Algorithms, Databases, and Tools, Mihai Pop
Education
|
|
|
