The HLA (Human Leukocyte Antigens) genes are well-documented targets of balancing selection, and variation at these loci is associated with many disease phenotypes. Variation in expression levels also influences disease susceptibility and resistance, but little information exists about the regulation and population-level patterns of expression due to the difficulty in mapping short reads to these highly polymorphic loci, and in accounting for the existence of several paralogues. We developed a computational pipeline to accurately estimate expression for HLA genes based on RNA-seq, improving both locus-level and allele-level estimates. First, reads are aligned to all known HLA sequences in order to infer HLA genotypes, then quantification of expression is carried out using a personalized index. We use simulations to show that expression estimates are not biased due to divergence from the reference genome. We applied our pipeline to GEUVADIS dataset, and compared the quantifications to those obtained with reference transcriptome, and found that a substantial portion of the variation captured by the HLA-personalized index in not captured by the standard index (23%). We describe the impact of the HLA-personalized approach on downstream analyses for seven HLA loci (HLA-A, HLA-B, HLA-C, HLA-DPB1, HLA-DQA1, HLA-DQB1, HLA- DRB1). Although the influence of the HLA-personalized approach is modest for eQTL mapping, the p-values and the causality of the eQTLs obtained are better than when the reference transcriptome is used. Finally, we integrate information on HLA-allele level expression with the eQTL findings to show that the HLA allele is an important layer of variation to understand HLA regulation.