Skip to main content
Premium Trial:

Request an Annual Quote

Capture Technique Enables Gene Comparisons Across Divergent Species


A technique that enables genes to be captured and sequenced across divergent species is enabling researchers to study evolution and could have biomedical applications.

"We want to target particular genes that have properties we're interested in across different organisms," Gavin Naylor, a professor at the College of Charleston's Hollings Marine Laboratory, told In Sequence.

Current gene panels tend to aim for specificity, said Naylor, which is important when you want to select for specific genes and only those genes in one organism. But, for instance, "a human cancer panel will not work on an orangutan," he said.

Thus, he and his research team designed a strategy that would be both "specific and general enough" to capture homologous genes across multiple organisms. The proof of principle was published in the June issue of BioTechniques.

To demonstrate the method, the team first looked for single-copy orthologous genes that were present in six divergent vertebrates — human, chicken, western clawed toad, green anole, zebrafish, and elephant shark genomes — identifying 1,449 candidate genes varying from 151 bases to 2,390 bases in size.

Next, they developed separate custom biotinylated RNA bait libraries for each species comprised of pooled 120-base baits for each target gene.

They then tested the capture on each of the six species to make sure that the baits could in fact capture the desired genes, and they also tested it on species within the same class of each of the six vertebrates to see if it could capture the genes from distantly related species within the same class.

So, within mammals, the team used the human genome to design the baits and attempted to capture the genes from an opossum. For birds, they used the chicken genome to create baits and tried to capture genes from the zebra finch. For reptiles, the baits designed from the green anole were used to capture genes from the painted turtle. For amphibians, baits from the western clawed toad were used to capture genes from the axolotl. And for fish, baits from the zebrafish were used to capture genes from the stickleback.

The range of divergence between the organisms was as great as 39 percent. Highly conserved sequences within the target genes were as great as 90 percent similar between species.

Additionally, they tested two different hybridization and washing strategies — both a standard approach and a more "relaxed" approach that allowed for more mismatches.

After performing the gene capture, 20 indexed samples were pooled and sequenced on Illumina's MiSeq with 150 base paired-ends. For each species, the researchers de novo assembled contigs.

Comparing the two different hybridization and washing approaches, the team found that under standard conditions, very few target sequences were retrieved except for the positive controls where both bait and target library were from the same species.

With the more relaxed approach though, there was an eight-fold increase in the number of target sequences captured. Additionally, doing two rounds of capture with the more relaxed approach increased the captured targets by another 68 percent.

The relaxed approach is "more promiscuous," explained Naylor, so it "binds to things that mismatch." However, adjusting the specificity of the chemistry there is a balancing act, he said. As the chemistry is relaxed, "there becomes a point that it binds to things that are not what we want, and we end up with rubbish."

The team found that with the optimized approach they were able to capture up to 1,159 genes of the 1,449 that they targeted between the chicken and the zebra finch.

They saw the worst results among the amphibians — capturing only 225 of the target genes when using the western clawed toad as bait to capture genes from the axolotl, which the authors speculated might have been due to repetitive elements within those genomes.

To examine what other factors aside from divergence between the bait and target species contributed to the efficacy of the capture technique, the researchers compared GC content, target sequence length, and chromosomal position between targets that were successfully captured and those that failed among the six positive controls where the bait and target were identical.

On average, the targets that were not captured tended to have higher GC content and shorter lengths than those that were captured.

Examining the technique within fish, the researchers targeted the same 1,449 from 13 different species. Within the same class, the technique demonstrated a greater efficiency, capturing between 1,004 genes in the worst case and 1,351 genes in the best case of the 1,449 targeted genes. For the positive control, all targets were captured. The results "show greater homogeneity in the efficacy of capture than was evident in the survey across vertebrate classes, confirming the expectation that the methods would be more consistent when applied to a denser taxonomic sample of more closely related lineages," the authors wrote.

Naylor said the protocol itself is relatively straightforward and based off the same principle as Southern blot technology, a method of detecting a specific sequence in a DNA sample.

While library preparation takes around one day and the gene capture around two days, there is a return on that investment in the amount of information generated, he said, which is about two orders of magnitude greater than what could be achieved with PCR.

"We're pulling out 1,000 genes for 20 animals at a time on the MiSeq," said Naylor. And, it is possible to scale up even further and do "all the genes in one pathway for 100 different animals in one HiSeq run," he said.

Naylor said that the group plans to use the technology in evolutionary biology studies, for instance to "explore the genetics that might be associated with particular genes that sprout in an evolutionary lineage," he said, such as the genes that allowed for fish to develop brains.

The same technique could also have broader applications within the biomedical industry as well, he said, for instance, in tracking down genes that may predispose someone to cancer, or make them more resistant to cancer.