Getting more from digital SNP data
Noureddine El Karoui, Wei Zhou, Alice S. Whittemore · Statistics in Medicine · 2006
The digital SNP method has been proposed for identifying loss of heterozygosity (LOH) in tumour tissue and correlating it with patients' clinical characteristics. The method evaluates a tumour's allelic count at a single nucleotide polymorphism (SNP) for which the patient's normal tissue is heterozygous. The count is used to classify the tumour as positive or negative for LOH, using the sequential probability ratio test (SPRT). However, the SPRT was not developed for analysing digital SNP experiments. When applied to digital SNP data, the SPRT has several anomalies that can result in both loss of data and tumour misclassification. The anomalies are caused by discrepancies between the design of digital SNP experiments and the setting for which SPRT was developed. We propose an alternative classification scheme based on the false discovery rate, and show that it outperforms the SPRT when applied to Digital SNP data.
Cited by 11 works. doi:10.1002/sim.2379
Authors
References
- Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing
- Digital PCR
- A systematic approach to the analysis of protein phosphorylation
- Sequential Analysis
- dChipSNP: significance curve and clustering of SNP-array-based loss-of-heterozygosity data
- Genome-wide Detection of Allelic Imbalance Using Human SNPs and High-density DNA Arrays
- Evidence that genetic instability occurs at an early stage of colorectal tumorigenesis.
- Small changes in expression affect predisposition to tumorigenesis
- Counting alleles to predict recurrence of early-stage colorectal cancers
- Assessment of Plasma DNA Levels, Allelic Imbalance, and CA 125 as Diagnostic Tests for Cancer