The goal of personalizing medical treatment on a genetic level is complicated by the human genome’s size: It contains about 22,000 genes that serve as blueprints for an even greater number of proteins needed to make cells function. Linking diseases and disorders to individual protein defects — and underlying genetic mutations — is a monumental and challenging task.
Researchers at the University of Maryland, Baltimore County (UMBC) are simplifying this task and accelerating progress toward personalized medicine by offering a new tool that catalogs these defects as they appear in the roughly 4,000 protein domains that recur in proteins.
The Domain Mapping of Disease Mutations (DMDM) database, bioinf.umbc.edu/dmdm, enables researchers to look for new links between diseases and therapeutic approaches by providing a map of the flaws that show up in what amounts to a “parts warehouse” for proteins.
“We are trying to change from a gene-centric perspective to a domain-centric view,” says Maricel Kann, an assistant professor in UMBC’s Department of Biological Sciences. “This approach will provide valuable perspective as researchers grapple with the vast amount of data that gene-sequencing technology has made available.”
Kann’s multidisciplinary research team — consisting of undergraduates and graduate students — described the database in an article published in August in the journal Bioinformatics. Updated this month, DMDM now tracks more than 120,000 mutations drawn from the three primary databases tallying research in genetic diseases, mutations and protein structures.
The DMDM database currently receives about 1,200 visits a week. Visitors can search by disease, gene, protein or domain. At each level, the database shows the location of known mutations. For domains, the database shows the number of mutations found in each location, with information about the associated proteins.
Domains are the basic functional units of proteins. Highly conserved throughout evolution, each domain gives a protein a new capability, such as the ability to bind to DNA or to catalyze a specific reaction.
To build the DMDM database, UMBC researchers downloaded the known disease-causing point mutations (a point mutation causes a change from one amino acid to another) from several public databases. They then mapped the mutations to positions within individual proteins and protein domains, finding that more than 75 percent of the disease mutations fall within protein domains.
This finding shows that domains are critical to the proper functioning of proteins: When a mutation falls within a protein domain, the protein’s function is often disrupted and disease can result.
As more and more individual genomes are sequenced, more new mutations will be identified. Researchers can use DMDM to check if a mutation falls within a domain to see if it is likely to disrupt protein function. Researchers can also check if a mutation in a certain position within the domain has previously been linked to disease. Ultimately, mutation “hotspots” within domains could be targeted by pharmaceutical companies for drug development.
A screenshot from the database is below: