Non-invasive Prenatal Paternity Testing (NIPPT) with Next Generation Sequencing

Description

Questions regarding paternity during pregnancy are common due to various reasons. Seeking answers and resolving such concerns can not only provide emotional benefits for the family but also alleviate the stress of the pregnant woman. Each case concerns a family. Reliable prenatal paternity is vulnerated and increase the risk of harm to the mother and fetus. The discovery of fetal cell-free DNA in maternal blood provided theoretical possibilities for the non-invasive prenatal paternity testing. Methods based on SNP or Y-chromosome have been developed. However, both methods have significant drawbacks. Y-chromosome based methods are gender dependent and they cannot handle the cases for female fetal. SNP based methods rely on the fact that if some alleles are 1) homozygous in maternal plasma and paternal DNA, and 2) heterozygous in fetal, then they can be applied for NIPPT. However, the observed heterozygous SNP markers are suspicious due to that the complex maternal plasma components and limited discriminative power of the SNP alleles. Moreover, such methods require high sequencing depth maternal DNA sample to separate fetal DNA component from maternal plasma. The price is approximately four times higher than traditional paternity testing. In this project, we propose a Bayesian-based statistical method. In this method, we combine SNP markers and STR markers to calculate the likelihood ratios to obtain the probability of whether the alleged man is the biological father of the fetus with DNA samples only from the maternal plasma and the alleged father. In the model, we handled uncertainty resulted from fluctuant fetal cfDNA, sequencing errors, and allelic dropout. In order to lower the cost, we will further examine the relationship between sequencing depth and accuracy to obtain the minimum sequencing coverage requirement.