Improving SNP Calling Accuracy with CRLMM
Author Information
Author(s): Lin Shin, Carvalho Benilton, Cutler David J, Arking Dan E, Chakravarti Aravinda, Irizarry Rafael A
Primary Institution: Johns Hopkins University
Hypothesis
Can the CRLMM algorithm provide more accurate SNP calls compared to existing methods?
Conclusion
The CRLMM algorithm is shown to be more accurate than the Affymetrix default programs for SNP calling.
Supporting Evidence
- CRLMM was trained on high-quality Affymetrix HapMap array sets.
- CRLMM outperformed BRLMM in accuracy across multiple datasets.
- CRLMM provides improved estimates of accuracy compared to BRLMM.
Takeaway
This study shows that a new method for reading DNA can make fewer mistakes than older methods, helping scientists understand genes better.
Methodology
The study involved extending and validating the CRLMM algorithm using high-quality HapMap datasets and comparing its performance against BRLMM.
Potential Biases
Potential bias exists due to the reliance on HapMap data for training the CRLMM algorithm.
Limitations
The study's results may be influenced by the training data derived from HapMap individuals, which could introduce bias.
Participant Demographics
The study utilized data from 269 individuals genotyped on Affymetrix 100K arrays.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website