Sequences of 95 humanMHChaplotypes reveal extreme coding variation in genes other than highly polymorphicHLA class IandII

The most polymorphic part of the human genome, theMHC,encodes over 160 proteins of diverse function. Half of them, including theHLA class IandIIgenes, are directly involved in immune responses. Consequently, theMHCregion strongly associates with numerous diseases and clinical therapies. Notoriously, theMHCregion has been intractable to high-throughput analysis at complete sequence resolution, and current reference haplotypes are inadequate for large-scale studies. To address these challenges, we developed a method that specifically captures and sequences the 4.8-MbpMHCregion from genomic DNA. For 95MHChomozygous cell lines we assembled, de novo, a set of high-fidelity contigs and a sequence scaffold, representing a mean 98% of the target region. Included are six alternativeMHCreference sequences of the human genome that we completed and refined. Characterization of the sequence and structural diversity of theMHCregion shows the approach accurately determines the sequences of the highly polymorphicHLA class IandHLA class IIgenes and the complex structural diversity of complement factorC4A/C4B. It has also uncovered extensive and unexpected diversity in otherMHCgenes; an example isMUC22, which encodes a lung mucin and exhibits more coding sequence alleles than anyHLA class IorIIgene studied here. More than 60% of the coding sequence alleles analyzed were previously uncharacterized. We have created a substantial database of robust referenceMHChaplotype sequences that will enable future population scale studies of this complicated and clinically important region of the human genome.