New weapon in an age-old battle
It’s a story as old as humankind. Boy (or girl) meets bug, a struggle for dominance ensues, and someone gets hurt.
When a pathogen that causes an infection can be identified, it can often be zapped with a readily available drug, like an antibiotic. And the genomic era has ushered in a new understanding of how even drug-resistant microbes can be targeted and tamped down. But science cannot always ferret out the identity of the bad guys that cause disease. In fact, researchers believe that many human diseases are caused by undiscovered pathogens, bugs so new that they have yet to be named and studied.
Now, next-generation genomic sequencing technology aims to give humanity better odds in this age-old battle.
Matthew Meyerson, a senior associate member of the Broad, and his colleagues have developed a genomic approach for discovering the genetic sequences of mystery microbes called sequence-based computational subtraction. Call it addition by subtraction: Researchers use high-throughput technologies at the Broad to sequence libraries of diseased tissues, then compare them with the sequence of the human genome, which was completed 10 years ago. Once the known DNA sequence of the human genome is filtered out, only the footprints of potential culprits remain, in the form of microbial sequences.
To streamline the process, Meyerson and his team developed a software package called PathSeq, which harnesses a clutch of 20 computers at the same time in parallel to crunch the data. PathSeq, which can be used on offsite computers – a technique known as cloud computing – can be launched and monitored on a simple laptop. With a single CPU, this task could take months. With parallel computing, it can take a day or two.
“With next-generation sequencing, everything has changed,” says Alex Kostic, a Ph.D. student in Meyerson’s lab and first author of a Nature Biotechnology paper on PathSeq published earlier this month. “It has now become so cheap that you can sequence everything that is there, and parse it out and see if there is a new pathogen. The idea is that if you do random, unbiased sequencing of human tissue and there’s a pathogen in that tissue, then you can just subtract out all the human sequences by comparison to the reference genome. What you’ll be left with is potential pathogens.”
Broad researchers are making PathSeq available to the scientific community at large. Although the algorithm needs to be finely tuned, researchers believe it will help them identify new pathogens and, ultimately, new treatments to combat them.
Aleksandar D. Kostic, Akinyemi I. Ojesina, Chandra Sekhar Pedamallu, Joonil Jung, Roel G.W. Verhaak, Gad Getz, and Matthew Meyerson. PathSeq: Software to identify or discover microbes by deep sequencing of human tissue. Nature Biotechnology, Vol. 29, 393-396, published online May 6, 2011 DOI:10.1038/nbt.1868