科学美国人60秒:Better Gut Microbiome Census through Computing

In recent years scientists have shown that the microbes that live in our guts play crucial roles in our lives. They’re involved in digestion, obesity, even mood. And a few can cause serious illness. So it would be a good idea to know the identities of the bacteria inside us. And yet, that info has been incomplete.

But now researchers have developed a technique to get a better census of the gut microbiome. And using the new system, the researchers have found that our microorganisms are even more diverse than we knew. The report is in the journal Nature Biotechnology. [Volodymyr Kuleshov et al, Synthetic long-read sequencing reveals intraspecies diversity in the human microbiome]

Currently, researchers analyze microbial diversity by taking a sample they hope includes the different kinds of bacteria in the gut. They then try to identify the different species by looking at their genomes. But they can only do that second-hand, by trying to piece together many short snippets of DNA—which can be confusing and inadequate when dealing with numerous different kinds of bacteria.

So geneticists at Stanford University got together with computer scientists to come up with a new approach. They used sophisticated computational techniques that enabled them to analyze much longer stretches of DNA—which included many genes that would be missed with the older system. For example, when they tested the gut microbiome from a healthy human male the old way, they found 127 different species. The new method applied to the same sample revealed the presence of an additional 51 species.   

The new approach could be particularly important for identifying and understanding disease-causing microbes. "When you assemble the whole genome, you have a really good idea of what pathogenic genes are present.” Michael Snyder, one of the study researchers. “So we think this technology is going to be extremely powerful for understanding the genetic basis of pathogenesis.”

For example, we all harbor benign strains of E. coli bacteria. But other strains can be toxic or even deadly—and they might be hard to investigate because they don’t grow easily in the lab. The new approach could look directly at the toxic strain’s genes to see how they functions. “And of course this will be really powerful then for treating humans in terms of what pathogenic genes might be present in the microorganisms they harbor.”

—Cynthia Graber