The blood transcriptional signature for active and latent tuberculosis

Abstract

Background: Although the incidence of tuberculosis (TB) has dropped substantially, it still is a serious threat to human health. And in recent years, the emergence of resistant bacilli and inadequate disease control and prevention has led to a significant rise in the global TB epidemic. It is known that the cause of TB is Mycobacterium tuberculosis infection. But it is not clear why some infected patients are active while others are latent.

Methods: We analyzed the blood gene expression profiles of 69 latent TB patients and 54 active pulmonary TB patients from GEO (Transcript Expression Omnibus) database.

Results: By applying minimal redundancy maximal relevance and incremental feature selection, we identified 24 signature genes which can predict the TB activation. The support vector machine predictor based on these 24 genes had a sensitivity of 0.907, specificity of 0.913, and accuracy of 0.911, respectively. Although they need to be validated in a large independent dataset, the biological analysis of these 24 genes showed great promise.

Conclusion: We found that cytokine production was a key process during TB activation and genes like CYBB, TSPO, CD36, and STAT1 worth further investigation.

Keywords: blood gene expression; incremental feature selection; minimal redundancy maximal relevance; support vector machine; tuberculosis.

Figure 1

The prediction performances for TB activation by using different numbers of signature genes. Notes: The x-axis is the number of genes in the gene set while y-axis is the prediction accuracy of the SVM classifier evaluated with LOOCV. The peak of the IFS curve had an accuracy of 0.919 when 51 genes were used. But when 24 genes were used, the accuracy has already become stable. Therefore, we choose these 24 genes as signature genes of TB activation. The sensitivity, specificity, and accuracy of the 24 signature genes for TB activeness prediction were 0.907, 0.913, and 0.911, respectively. Abbreviations: LOOCV, leave-one out-cross validation; IFS, incremental feature selection; SVM, support vector machine; TB, tuberculosis.

Figure 2

The heatmap of the 24 signature genes latent and active TB patients. Notes: The rows represent genes while the columns represent patients. The green and red columns represent latent and active TB patients, respectively. It can be seen that the latent and active TB patients were clustered into different groups. Abbreviation: TB, tuberculosis.