Accuracy and differential bias in copy number measurement of CCL3L1 in association studies with three auto-immune disorders
1 Centre for Genetics and Genomics and School of Biology, University of Nottingham, Nottingham NG7 2UH, UK
2 Division of Genetics and Molecular Medicine, King's College London School of Medicine, 8th Floor Guy's Tower, Guy's Hospital, London SE1 9RT, UK
3 Department of Dermatology, Radboud University Nijmegen Medical Centre, Nijmegen Centre for Molecular Life Sciences, 6525 GA Nijmegen, The Netherlands
BMC Genomics 2011, 12:418 doi:10.1186/1471-2164-12-418Published: 18 August 2011
Additional file 1:
Figure S1. The distribution of unrounded copy number values for 192 ECACC samples typed with the "CCL3C" system, with discernable peaks around the integers, comparable to the original "CCL3A" distribution (figure 2a in Walker et al. 2009 ). The "CCL3A" method had an overall standard deviation of 0.087, whereas the modified "CCL3C" method had a standard deviation of 0.058 for the full dataset.
Format: TIFF Size: 46KB Download file
Additional file 2:
Figure S2. Histograms of the cases and control samples for Crohn's disease samples (a), rheumatoid arthritis (b), and psoriasis (c), with cases in black and controls in white. The histograms show no significant difference between the cases and controls for all datasets.
Format: TIFF Size: 123KB Download file
Additional file 3:
Figure S3. Q-Q plots of the control samples fitted to a normal distribution for copy numbers of 1 (n = 49) (control1) (a), 2 (n = 149) (control2) (b) and 3 (n = 45) (control3) (c).
Format: TIFF Size: 382KB Download file