Table 4

QRT-PCR validation and phenotype prediction



qRT-PCR results: MFS/UC ratio (Wilcoxon p-value)



Gene symbol
Array ratio
Array subjects
New subjects
All subjects

VDR
1.42-1
4.59-1
(9.5e-5)
1.25
(8.1e-1)
2.53-1
(7.3e-3)
FBN1
1.89-1
1.47-1
(5.2e-2)
2.71
(3.3e-9)
N/A

INHBA
1.61-1
2.20
(1.1e-1)
1.06-1
(6.7e-1)
1.05
(5.0e-1)
ELN
3.34-1
12.53-1
(2.4e-3)
9.68-1
(2.3e-4)
15.04-1
(1.6e-8)
COL1A2
1.80-1
1.24-1
(1.1e-1)
2.41-1
(2.3e-8)
2.27-1
(1.4e-4)
PCOLCE
1.26-1
1.27-1
(5.8e-2)
2.02
(4.6e-6)
N/A

PLOD2
1.22
3.01
(3.8e-4)
1.69
(8.2e-3)
1.85
(2.7e-4)
PDCD10
1.15
1.98
(1.9e-6)
1.65
(4.8e-3)
1.41
(5.0e-5)
PTGES
1.72
3.42-1
(5.8e-4)
1.88
(9.3e-3)
1.68-1
(6.8e-2)
LMO7
1.13
2.01
(1.0e-3)
4.67
(3.0e-6)
2.77
(1.7e-6)








Overall Significance
Array subjects
New subjects
All subjects


5
(6.4e-5)
6
(2.8e-6)
6
(2.8e-6)

Column 2: average MFS/UC ratio from array experiments. Remaining columns: ratios of geometric mean transcript abundances of MFS affected vs unaffected control for three subject groups, and Wilcoxon p values for the null hypothesis of no between group differences. The three groups are: (1) "Array Subjects": 30 of the original 36 array subjects (16 known FBN1 mutations and 14 unaffected controls); (2) "New Subjects": 32 new subjects including 16 MFS based on clinical criteria and 16 unaffected controls; and (3) "All Subjects": 42 MFS affected compared to 32 unaffected (all of the above groups plus 10 additional characterized FBN1 mutations and 2 additional controls). (See Table 1 for subject details.) Bold: array validation or significant p value (p < 0.05). Italic: significant p-value but difference opposite to the expected direction. N/A: see text. The bottom section of the table provides a summative assessment: the (binomial) p-value of the observed number of significant qRT-PCR validations (p < 0.05) under the null hypothesis of no between-group differences.

Yao et al. BMC Genomics 2007 8:319   doi:10.1186/1471-2164-8-319