COFFEE: control‐free noninvasive fetal chromosomal examination using maternal plasma DNA

Abstract Objective The aim of this study is to develop an approach for analyzing plasma DNA sequencing data for noninvasive fetal chromosomal aneuploidy testing that does not require the comparison with control samples or a series of selected genomic regions. Results We developed the control‐free noninvasive fetal chromosomal examination (COFFEE) algorithm by utilizing the size differences between the fetally derived and maternally derived DNA molecules in maternal plasma. We applied COFFEE on three datasets generated in different experimental settings. COFFEE showed 100% accuracy in trisomy 21 testing on these datasets. In contrast, samples analyzed using an existing control‐based z ‐score method would introduce a false‐positive result because of batch‐to‐batch variation, when the tested samples were analyzed using control samples from other batches. We believe that COFFEE is useful for enhancing the cost‐effectiveness of noninvasive fetal chromosomal aneuploidy testing particularly in laboratories with small caseloads. Source code and testing datasets for COFFEE are available for download at http://www.cuhk.edu.hk/med/cpy/Research/COFFEE/ . Conclusion Control‐free noninvasive fetal chromosomal examination is demonstrated to be a versatile data analysis approach and could enhance the application of noninvasive fetal chromosomal aneuploidy detection. © 2017 John Wiley & Sons, Ltd.