NEW YORK (GenomeWeb) – A BGI-Shenzhen-led team has demonstrated that it is possible to profile population genetics, viral infection patterns, and more using maternal DNA profiles obtained by non-invasive prenatal testing (NIPT).
For a paper published in Cell today, the researchers used new statistical approaches to assess genome sequences generated from maternal DNA in NIPTs for more than 140,000 Chinese women. Along with insights into China's population structure and history, they showed that this genetic data could be used for finding associations with traits such as height, body mass index, maternal age at first birth, or twin pregnancy, and for profiling viral infection prevalence and susceptibility in the pregnant women.
"We show that we can solve many classical population genetic and medical problems by analyzing NIPT data," co-first author Siyang Liu, a researcher affiliated with BGI-Shenzhen and the University of Copenhagen, said in a statement. "As the testing is going to be undertaken by millions worldwide, it's more and more important to establish collaborations so that it is more possible to obtain the informed consent of patients and to properly manage medical records."
Chinese women are thought to have received some 70 percent of the more than 10 million NIPTs done so far, the team noted, prompting speculation that this large set of samples could offer a window into broader population patterns and genetic associations.
"Although non-invasive prenatal testing is low-pass sequencing, capturing approximately [6 percent to 10 percent] of the whole genome randomly, there's still a chance that using this data with a large population size will help us to have a much broader vision of what the Chinese genetic population looks like," co-senior author Xun Xu, executive vice president at BGI and director at BGI-Shenzhen, said in a statement.
To explore that possibility, Liu, Xu, and colleagues enrolled 141,431 NIPT recipients from dozens of administrative divisions in China, using Illumina HiSeq 2000 instruments to produce between 0.06-fold and 0.1-fold sequence depth per person. They also sequenced samples from a subset of 40 participants to deeper, 15-fold average coverage, using variants in these genomes to bolster genotype imputations in the broader group.
Based on information at millions of genotyped or imputed variants across the full set of participants, including representatives from three dozen ethnic minorities in the country, the team started teasing out population structure, history, and the genetic adaptations found in China.
The results revealed a north-to-south genetic gradient within the Han Chinese individuals, despite overall genetic homogeneity in this population. In contrast, individuals from ethnic minority groups such as the Kazakh, Uyghur, or Mongol populations showed more pronounced genetic differences depending on their location.
Along with analyses looking at genetic ties to populations from other parts of the world, the researchers used allele frequency patterns to retrace migrations within China, including historical migrations into Beijing and more recent internal migrations due to government interventions and labor movement.
They also searched for signs of selection in the Chinese populations, and showed that the available genetic data may be harnessed for association studies. In the latter analysis, for example, the tracked-down loci linked to everything from height and weight to pregnancy-related traits such as maternal age at first birth and twin pregnancy.
The twin pregnancy analysis, based on data for almost 138,000 women with ultrasound data available, led to a variant that influences the expression of a thyroid-related gene expression and was previously linked to hyperthyroidism in Europeans. The team noted that this result appears biologically relevant since enhanced thyroid hormone activity typically coincides with lower-than-normal thyroid-stimulating hormone levels — a trait associated with twin pregnancy in the past.
"Our results illustrate the utility of NIPT data for medical genetic studies, particularly for understanding traits related to fertility and pregnancy," the authors wrote. "Furthermore, the availability of large samples of shotgun DNA sequencing from blood opens up new avenues for investigating hypotheses regarding interactions between viruses and host DNA genetic variability."
In addition to getting a glimpse at the viruses circulating in maternal blood in general, the researchers identified genetic ties to circulating herpesvirus 6A/B (HHV-6A/B) levels and a variant falling near a gene called MOV10L1, which codes for a helicase enzyme protein involved in both platelet distribution and PIWI-interacting RNA processing functions during spermatogenesis.
"We suspect that the PIWI-interacting RNA represses HHV-6A/B integration, and polymorphisms in this gene allow for more efficient integration of HHV-6A/B during spermatogenesis," the authors wrote.