NEW YORK (GenomeWeb) – Researchers have generated two new great ape genome assemblies that they say are better able to help scientists spot genomic differences between humans and their closest relatives.
Previous great ape genome assemblies suffered from thousands of gaps, and using the human reference genome to guide these assemblies may have led them to be "humanized" and may have masked structural and other differences, said the researchers, who were led by the University of Washington's Evan Eichler.
To address those issues, they used both long-read sequencing and cDNA sequencing to generate new chimpanzee and orangutan genome assemblies without relying on the human genome reference. By comparing these to two long-read de novo human genome assemblies and a gorilla genome assembly, Eichler and his colleagues were then able to tease out lineage-specific differences, as they reported in Science today.
"Our goal," Eichler said in a statement, "is to generate multiple ape genomes with as high quality as the human genome. Only then will we be able to truly understand the genetic differences that make us uniquely human."
The researchers sequenced the genomes of two human, one chimpanzee, and one orangutan to more than 65-fold coverage using Pacific Biosciences' long-read sequencing approach.
About 93 percent each of the chimpanzee and orangutan bases could be incorporated into chromosome-level scaffolds, without using the human reference genome as a guide, Eichler and his colleagues reported. They estimated that these assemblies improved the contiguity of the chimpanzee and orangutan genomes by a respective 32- and 533-fold.
The researchers likewise generated long-read transcriptome sequencing data from induced pluripotent stem cells they generated from chimpanzee, orangutan, and gorilla to develop de novo gene models.
In a five-way genome-wide multiple sequence alignment — of the human, chimpanzee, orangutan, and previously sequenced gorilla genomes — the researchers found that 83 percent of the ape genomes could be compared.
For repeat elements, which the researchers noted often misassembled, they focused on the endogenous retrovirus PtERV1, which is found in chimpanzee and gorilla but not in orangutan and human genomes. They found that the PtERV1 integrations are largely not orthologous, are biased against genes, and are in the anti-sense orientation, which they noted was consistent with the effects of purifying selection.
They also homed in on what they termed the "source PtERV1"— the one orthologous chimpanzee and gorilla PtERV1 element. It likely didn't spread to the human lineage due to incomplete lineage sorting, they noted, and wasn't picked up in previous studies as it is found in a repeat-rich site.
The researchers also identified more than 614,000 deletions, insertions, and inversions among the ape lineages. At the same time, they uncovered 17,789 fixed human-specific structural variants. Annotating these fhSVs against chimpanzee and human gene models, the researchers found that these variants disrupt nearly 650 regulatory regions near almost 480 genes.
One fhSV, for instance, deletes part of the FADs1 and FADs2 genes, which are involved in fatty acid biosynthesis, while other fhSVs affect the cell cycle genes WEE1 and CDC25C, leading to additional cell divisions. Additional cell divisions, the researchers noted, are thought to be involved in the increased number of cortical neurons observed in humans. Together, these changes could account for some of the differences between great apes and humans.
The researchers also investigated differences in gene expression in chimpanzees and humans using cerebral organoids. They found hundreds of genes that are up- or downregulated in human radial glial and excitatory neurons, as compared to chimpanzees. Of these, 252 radial glia genes and 123 excitatory neuronal genes had annotated fhSVs, and the researchers noted that downregulated genes in humans tend to be enriched for fhSVs.
This, Eichler and his colleagues said, is consistent with the "less is more" hypothesis posited by Maynard Olson in the 1990s that said that the loss of functional elements accounts for key parts of human evolution.