Table 2.
Replication Study of Candidate Genes
| Affymetrix Results | CodeLink Results | ||||||||
| Gene Symbol | HXB23 | HXB26 | Ratio | p-value | HXB23 | HXB26 | Ratio | p-value | Match Direction |
| Afap1l1 | 6.44 | 6.69 | 0.84 | 0.0344 | 1.22 | 1.58 | 0.78 | 0.0199 | Yes |
| Cckbr | 7.92 | 8.31 | 0.76 | 0.0063 | 2.25 | 2.84 | 0.66 | 0.0110 | Yes |
| P2rx4 | 8.45 | 7.52 | 1.90 | 0.0122 | 2.53 | 1.50 | 2.04 | <.0001 | Yes |
| Tnk2 | 8.80 | 8.68 | 1.09 | 0.3473 | 4.17 | 3.59 | 1.50 | 0.0214 | Yes |
| Tubb6 | 6.92 | 7.00 | 0.94 | 0.5960 | 0.30 | 0.69 | 0.76 | 0.0400 | Yes |
| Ptpre | 7.70 | 7.63 | 1.05 | 0.2589 | 1.93 | 0.77 | 2.23 | <.0001 | Yes |
| Mogat2 | 5.26 | 5.10 | 1.12 | 0.3124 | 0.63 | -0.63 | 2.39 | 0.0005 | Yes |
| Nek3 | 6.05 | 5.90 | 1.10 | 0.1681 | -0.36 | -0.54 | 1.13 | 0.0388 | Yes |
| Tmem2 | 7.25 | 7.32 | 0.95 | 0.3860 | 2.47 | 2.82 | 0.78 | 0.0086 | Yes |
| Mc4r | 6.01 | 5.82 | 0.87 | 0.0360 | -0.39 | -0.26 | 0.91 | 0.2058 | Yes |
| F2rl1 | 6.16 | 6.26 | 0.94 | 0.3674 | -2.10 | -1.72 | 0.77 | 0.3943 | Yes |
| Abat | 10.52 | 10.35 | 1.13 | 0.0669 | 4.71 | 4.66 | 1.04 | 0.4464 | Yes |
| Mboat2 | 8.73 | 8.46 | 1.21 | 0.0797 | 0.95 | 1.31 | 0.78 | 0.0051 | No |
| Coq7 | 8.22 | 8.02 | 1.15 | 0.2095 | 2.60 | 3.76 | 0.45 | <.0001 | No |
| Dynlrb1 | 11.27 | 11.23 | 1.03 | 0.7064 | 5.96 | 6.00 | 0.97 | 0.4726 | No |
| Tpst1 | 7.70 | 7.56 | 1.10 | 0.1642 | 3.12 | 3.77 | 0.64 | 0.0004 | No |
| Fgfr2 | 9.44 | 9.41 | 1.02 | 0.7038 | -1.10 | 4.25 | 0.02 | <.0001 | No |
| Pop5 | 7.53 | 7.55 | 0.98 | 0.9056 | 2.58 | 2.45 | 1.09 | 0.1509 | No |
| Rab35 | 8.91 | 8.87 | 1.03 | 0.7719 | 4.18 | 4.28 | 0.93 | 0.0107 | No |
| Spire1 | 9.78 | 9.74 | 1.03 | 0.3554 | 1.50 | 3.26 | 0.30 | <.0001 | No |
Expression levels of the identified candidate genes were determined in brains (n = 5 per strain) of rats from two of the RI strains, HXB RI 26 (high alcohol consumption) and HXB RI 23 (low alcohol consumption) using CodeLink Rat Whole Genome Bioarrays and Affymetrix GeneChip Rat Gene 1.0ST arrays. These strains were chosen because the expression levels of all candidate genes (on CodeLink arrays) that were significantly correlated with alcohol consumption when all the RI strains were analyzed, showed an appropriate direction of differential expression in these two strains compared to the correlation analysis, even if the difference in expression between these 2 particular strains was not statistically significant. Details on analysis, normalization and statistical comparisons are provided in the text. Values represent mean of the log base 2 transformed expression data. "Ratio" is the HXB23/HXB26 ratio of the unlogged intensity values. P values are from the two-sample t-test, and "match direction" refers to the comparison between the two array experiments.