Identifying differential exon splicing using linear models and correlation coefficients

BMC Bioinformatics

Table 3 Correlation Coefficients

	ITSN	IMMT	KIF1B	SLC25A	WNK1
breast vs. non-breast	0.9889	0.9734	0.9473	0.9770	0.9709
cereb. vs. non-cereb.	0.5605	0.9823	0.7727	0.9661	0.9764
heart vs. non-heart	0.9711	0.8616	0.6756	0.9302	0.9410
kidney vs. non-kidney	0.9656	0.9901	0.9215	0.9799	0.8768
liver vs. non-liver	0.9766	0.9452	0.9197	0.9875	0.9793
muscle vs. non-muscle	0.9799	0.9834	0.8285	0.9412	0.9771
panc. vs. non-panc.	0.9158	0.9772	0.9333	0.9747	0.9427
prost. vs. non-prost	0.9807	0.9940	0.9655	0.9786	0.9431
spleen vs. non-spleen	0.9852	0.9790	0.8928	0.9739	0.9769
testes vs. non-testes	0.9864	0.9960	0.8198	0.9717	0.9827
thyroid vs. non-thyroid	0.9889	0.9734	0.9473	0.9770	0.9709
z-score between highest and lowest coefficients	9.71	4.42	6.68	2.76	5.05

Pearson correlation coefficients of the 5 genes (rounded to 4 significant digits) calculated for all eleven tissue comparisons using the unfiltered dataset. Values in bold indicate that the gene had a significant p-value (corrected p-value < 0.0001) in that SI/LIMMA group comparison. The highest and lowest correlation coefficients for each gene were found to be significantly different (all had z-scores greater than 2).

ISSN: 1471-2105