Comparison of variations detection between whole-genome amplification methods used in single-cell resequencing.

Han Liang,Liao Chang,Bo Li,Xun Xu,Ting Wang,Shirley Tsang,Michael Dean,Fuqiang Li,Niharika Bedekar,Liying Song ,Shi Xiao-Li ,Huawei Wu ,Guoyun Xie,Geng Li ,Huanming Yang,Na Lü ,Ling Wang,Yong Hou,Xiuqing Zhang,Jianghao Chen,Runze Jiang,Jun Wang,Guibo Li,Xiaolong Zhang,Yingrui Li,Kui Wu

doi:10.1186/s13742-015-0068-3

Abstract

BackgroundSingle-cell resequencing (SCRS) provides many biomedical advances in variations detection at the single-cell level, but it currently relies on whole genome amplification (WGA). Three methods are commonly used for WGA: multiple displacement amplification (MDA), degenerate-oligonucleotide-primed PCR (DOP-PCR) and multiple annealing and looping-based amplification cycles (MALBAC). However, a comprehensive comparison of variations detection performance between these WGA methods has not yet been performed.ResultsWe systematically compared the advantages and disadvantages of different WGA methods, focusing particularly on variations detection. Low-coverage whole-genome sequencing revealed that DOP-PCR had the highest duplication ratio, but an even read distribution and the best reproducibility and accuracy for detection of copy-number variations (CNVs). However, MDA had significantly higher genome recovery sensitivity (~84 %) than DOP-PCR (~6 %) and MALBAC (~52 %) at high sequencing depth. MALBAC and MDA had comparable single-nucleotide variations detection efficiency, false-positive ratio, and allele drop-out ratio. We further demonstrated that SCRS data amplified by either MDA or MALBAC from a gastric cancer cell line could accurately detect gastric cancer CNVs with comparable sensitivity and specificity, including amplifications of 12p11.22 (KRAS) and 9p24.1 (JAK2, CD274, and PDCD1LG2).ConclusionsOur findings provide a comprehensive comparison of variations detection performance using SCRS amplified by different WGA methods. It will guide researchers to determine which WGA method is best suited to individual experimental needs at single-cell level.Electronic supplementary materialThe online version of this article (doi:10.1186/s13742-015-0068-3) contains supplementary material, which is available to authorized users.

Highlights

Single-cell resequencing (SCRS) provides many biomedical advances in variations detection at the single-cell level, but it currently relies on whole genome amplification (WGA)
Three WGA strategies are widely used for SCRS: degenerate-oligonucleotide-primed polymerase chain reaction (DOP-PCR) [11, 12], multiple displacement amplification (MDA) [13,14,15], and a combination of displacement pre-amplification and PCR amplification
We found that SCRS data from MDA had comparable single-nucleotide variations (SNVs) detection accuracy and copy-number variations (CNVs) detection accuracy to that of multiple annealing and looping-based amplification cycles (MALBAC)

Summary

Introduction

Single-cell resequencing (SCRS) provides many biomedical advances in variations detection at the single-cell level, but it currently relies on whole genome amplification (WGA). Three WGA strategies are widely used for SCRS: degenerate-oligonucleotide-primed polymerase chain reaction (DOP-PCR) [11, 12], multiple displacement amplification (MDA) [13,14,15], and a combination of displacement pre-amplification and PCR amplification (marketed as PicoPlex kit by Rubicon Genomics [16, 17], MALBAC kit by Yikon Genomics [7, 18, 19]). These three WGA strategies differ in the enzymes used and in the experimental protocol design, which may yield different performances and biases, allowing for different specific applications. Voet et al reported the variations detection performance comparison using human cell line and blastomeres amplified by MDA and PicoPlex WGA [22]

Methods

Results

Conclusion