Can the expression of histocompatibility antigen be changed by lithium?

Combining gene expression, demographic and clinical data in modeling disease: a case study of bipolar disorder and schizophrenia

Background

This paper presents a retrospective statistical study on the newly-released data set by the Stanley Neuropathology Consortium on gene expression in bipolar disorder and schizophrenia. This data set contains gene expression data as well as limited demographic and clinical data for each subject. Previous studies using statistical classification or machine learning algorithms have focused on gene expression data only. The present paper investigates if such techniques can benefit from including demographic and clinical data.

Results

We compare six classification algorithms: support vector machines (SVMs), nearest shrunken centroids, decision trees, ensemble of voters, naïve Bayes, and nearest neighbor. SVMs outperform the other algorithms. Using expression data only, they yield an area under the ROC curve of 0.92 for bipolar disorder versus control, and 0.91 for schizophrenia versus control. By including demographic and clinical data, classification performance improves to 0.97 and 0.94 respectively.

Conclusion

This paper demonstrates that SVMs can distinguish bipolar disorder and schizophrenia from normal control at a very high rate. Moreover, it shows that classification performance improves by including demographic and clinical data. We also found that some variables in this data set, such as alcohol and drug use, are strongly associated to the diseases. These variables may affect gene expression and make it more difficult to identify genes that are directly associated to the diseases. Stratification can correct for such variables, but we show that this reduces the power of the statistical methods.

Neither single-marker nor haplotype analyses support an association between genetic variation near NOTCH4 and bipolar disorder

Markers near the NOTCH4 locus on chromosome 6p21.3 have been reported to be associated with schizophrenia in some studies. Since schizophrenia and bipolar affective disorder (BPAD) may share genetic determinants, we tested markers in and near NOTCH4 in a sample of 153 parent-offspring triads ascertained through a sibling pair with BPAD for evidence of association. This sample would have 80% power to detect an association at or above a genotype relative risk of 2.4 at the 10(-7) level of significance. In addition to the two markers previously showing the most significant association with schizophrenia, three additional nearby markers were studied. The five markers were genotyped using validated methods. Both single-marker and 3-marker haplotype data was analyzed using family-based association methods. No genome-wide significant association was detected between any of the five SNP-markers and BPAD in this sample. One marker showed nominal evidence of association (P = 0.049), but this evidence was not supported by haplotype analyses including nearby flanking markers or by case-control analysis using 93 Caucasian controls. These results do not support an association between genetic variation near NOTCH4 and BPAD in this sample.