Differences between the genomes of lymphoblastoid cell lines and blood-derived samples

Abstract:

Lymphoblastoid cell lines (LCLs) represent a convenient research tool for expanding the amount of biologic material available from an individual. LCLs are commonly used as reference materials, most notably from the Genome in a Bottle Consortium. However, the question remains how faithfully LCL-derived genome assemblies represent the germline genome of the donor individual as compared to the genome assemblies derived from peripheral blood mononuclear cells. We present an in-depth comparison of a large collection of LCL- and peripheral blood mononuclear cell-derived genomes in terms of distributions of coverage and copy number alterations. We found significant differences in the depth of coverage and copy number calls, which may be driven by differential replication timing. Importantly, these copy number changes preferentially affect regions closer to genes and with higher GC content. This suggests that genomic studies based on LCLs may display locus-specific biases, and that conclusions based on analysis of depth of coverage and copy number variation may require further scrutiny.

Keywords::

• genomics
• whole-genome sequencing
• viral transformation
• copy number changes
• bioinformatics

Acknowledgments

The authors wish to thank Max Robinson and Vivek Rai for helpful discussions, Denise Mauldin and Chris Witwer for technical support, the Center for Systems Biology (P50GM075647) and the many individuals whose genomes were sequenced. Genome data production was supported by the University of Luxembourg-Institute for Systems Biology Program.

Disclosure

The authors report no conflicts of interest in this work.