Abstract:
β-thalassemia/Hb E disease is one of the most common thalassemias in
Southeast Asia. Patients with compound heterozygote for Hb E and β mutant alleles
display remarkable variability in clinical expression, ranging from nearly
asymptomatic to severe, transfusion-dependent disease. It is believed that additional
genetic factors modifying disease severity may account for this variability. Data
mining techniques were applied to a previous genome-wide association study of β0-
thalassemia/Hb E patients with either mild or severe disease severity. The strongest
association to disease severity was observed in the β-globin cluster. High density-SNP
mapping of β-globin cluster suggested two distinct linkage disequilibrium (LD)
blocks. Within large LD block, 52% of patients showed high heterozygosity. Three
tagSNPs of β-globin cluster were identified as biomarkers for prediction of disease
severity with 81% classification accuracy. A novel algorithm for gene-gene interaction
analysis called parallel haplotype configuration reduction (pHCR) was proposed.
Gene-gene interaction analysis suggested the interaction between DEPDC2, PIK3C2A
and HBS1L which was previously reported as quantitative trait loci (QTLs) associated
to Hb F levels. Lastly, the expression of α-hemoglobin stabilizing protein (AHSP), as
a candidate genetic modifier gene, was explored. No association between AHSP
expression and disease severity was observed. However, the result suggested that the
modulation of AHSP expression level was due to several polymorphisms within the
AHSP gene.