- Email: [email protected]
P059
92
P058
Abstracts / Human Immunology 75 (2014) 50–141
THE CHARACTERISTICS OF KIR2DL1 ALLELES POLYMORPHISM AND RECOGNITION HLAC LIGAND IN THE CHINESE HAN POPULATION. Jun He, Miao Wang, Xiaojing Bao, Jiang Zhang. The First Affiliated Hospital of Soochow University, Suzhou, China. Aim: To study the distributed characteristics of KIR2DL1 alleles frequencies and the recognition HLA-C ligand in the Chinese Han population. Methods: The 111 patients and 116 donors from CMDP were performed the KIR2DL1 high-resolution typing and KIR genotyping using sequence-based testing (SBT) and PCR-SSP methods. Results: A total of 224 individuals with KIR2DL1 locus was predominantly observed and accounted for 98.68% (224/227). There were 3 difference KIR2DL1 alleles, include of KIR2DL1⁄00302, ⁄00201 and ⁄00401 alleles polymorphism. The most commonly of phenotype frequency were observed KIR2DL1⁄00302(84.82%, 380/448), KIR2DL1⁄00201 (12.05%, 54/448) and KIR2DL1⁄00401 (3.13%, 14/448), present at allele genetype frequencies of 61.04%, 6.22% and 1.58%. The allele homozygotic type of KIR2DL1⁄00302 and KIR2DL1⁄00302 was the most frequency in the six gene complex of KIR2DL1 allele by high resolution typing. The allele heterozygous type of KIR2DL1⁄00302 and KIR2DL1⁄00401 had present statistically significant difference in haplotypes A/A and B/x(P = 0.001), and KIR2DL1⁄00401 was lacked all A/A haplotype. The KIR2DL1⁄00302 and KIR2DL1⁄00201 allele had significant positive associations between different KIR pairs of KIR2DS1,KIR2DS4,KIR2DS5,KIR2DL3 and KIR3DL1/S1 using linkage disequilibrium analysis (P 6 0.001), respectively. In the receptor-ligand of KIR/HLA model after allo-HSCT, KIR2DL1⁄00302 alleles and their HLA-C2 group ligands had correlation. KIR2DL1⁄00302 and HLA-C⁄06:02 was the most common of combination ligand model, but KIR2DL1⁄00302 and HLA-C⁄01:02 was the most frequent mismatch ligand model that it could be develop of NK cell-induced alloreactivity, meanwhile there was statistically significant difference of frequency distribution (P < 0.05). Conclusion: The KIR2DL1⁄00302 is the most frequency allele in Chinese Han population. The KIR2DL1 high resolution typing would be beneficial for predicting donor NK cells alloactivity after hematopoietic stem cell transplantation and selecting suitable donors.
P059
META ANALYSIS OF SEQUENCE-BASED HLA TYPING APPROACHES. Szilveszter Juhos, Krisztina Rigo, Gyorgy Horvath, Tim Hague. Omixon Biocomputing Ltd, Budapest, Hungary. Aim: Sanger nucleotide traces are used in many laboratories and clinics to get accurate and high resolution HLA types, and Next-Generation Sequencing (NGS) reads are just starting to be used for this purpose. Having both capillary and NGS reads from HLA genes makes it possible to study these different approaches to HLA Typing. We present two different NGS HLA typing analysis methods and compare their results to capillary trace based typing. Methods: Eight human samples, with amplicons covering HLA-A,B,C and HLA-DRB1 were analyzed. These amplicons were amplified and sequenced by Sanger capillary sequencing, Illumina MiSeq and IonTorrent instruments. The resulting NGS reads were typing by two different HLA typing algorithms. The first algorithm is a probability-based approach which distributes reads among allele candidates and selects the most likely ones using a comparison of coverage statistics. The second algorithm is a consensus-based approach that aims to deliver phased, positioned consensus sequences and then associate these with the alleles that are supported by the most reads from the provided sample data. Results: Capillary reads can be ambiguous, and covering long genomic parts with Sanger traces can be difficult and impractical. However, Sanger traces are the longest available ones with high quality. Statistics based typing can produce reliable results, but can lose local information (i.e. important SNPs). In addition, phase ambiguities, reads from homologous regions and PCR cross-overs may render statistical typing less reliable. Consensus based typing is less sensitive to homologous or PCR cross-over reads, and can deliver long consensus sequence stretches. This makes possible to discover new alleles or extend existing incomplete ones. Conclusion: NGS brings the hope of covering whole genes, but to assign reads among homologous genes and similar alleles has its own challenges. Comparing the three typing methods reveals differences in dealing with phase ambiguity, novel allele handling and quality control. NGS makes it possible to fully resolve phase and obtain long consensus sequences.
P058
Abstracts / Human Immunology 75 (2014) 50–141
THE CHARACTERISTICS OF KIR2DL1 ALLELES POLYMORPHISM AND RECOGNITION HLAC LIGAND IN THE CHINESE HAN POPULATION. Jun He, Miao Wang, Xiaojing Bao, Jiang Zhang. The First Affiliated Hospital of Soochow University, Suzhou, China. Aim: To study the distributed characteristics of KIR2DL1 alleles frequencies and the recognition HLA-C ligand in the Chinese Han population. Methods: The 111 patients and 116 donors from CMDP were performed the KIR2DL1 high-resolution typing and KIR genotyping using sequence-based testing (SBT) and PCR-SSP methods. Results: A total of 224 individuals with KIR2DL1 locus was predominantly observed and accounted for 98.68% (224/227). There were 3 difference KIR2DL1 alleles, include of KIR2DL1⁄00302, ⁄00201 and ⁄00401 alleles polymorphism. The most commonly of phenotype frequency were observed KIR2DL1⁄00302(84.82%, 380/448), KIR2DL1⁄00201 (12.05%, 54/448) and KIR2DL1⁄00401 (3.13%, 14/448), present at allele genetype frequencies of 61.04%, 6.22% and 1.58%. The allele homozygotic type of KIR2DL1⁄00302 and KIR2DL1⁄00302 was the most frequency in the six gene complex of KIR2DL1 allele by high resolution typing. The allele heterozygous type of KIR2DL1⁄00302 and KIR2DL1⁄00401 had present statistically significant difference in haplotypes A/A and B/x(P = 0.001), and KIR2DL1⁄00401 was lacked all A/A haplotype. The KIR2DL1⁄00302 and KIR2DL1⁄00201 allele had significant positive associations between different KIR pairs of KIR2DS1,KIR2DS4,KIR2DS5,KIR2DL3 and KIR3DL1/S1 using linkage disequilibrium analysis (P 6 0.001), respectively. In the receptor-ligand of KIR/HLA model after allo-HSCT, KIR2DL1⁄00302 alleles and their HLA-C2 group ligands had correlation. KIR2DL1⁄00302 and HLA-C⁄06:02 was the most common of combination ligand model, but KIR2DL1⁄00302 and HLA-C⁄01:02 was the most frequent mismatch ligand model that it could be develop of NK cell-induced alloreactivity, meanwhile there was statistically significant difference of frequency distribution (P < 0.05). Conclusion: The KIR2DL1⁄00302 is the most frequency allele in Chinese Han population. The KIR2DL1 high resolution typing would be beneficial for predicting donor NK cells alloactivity after hematopoietic stem cell transplantation and selecting suitable donors.
P059
META ANALYSIS OF SEQUENCE-BASED HLA TYPING APPROACHES. Szilveszter Juhos, Krisztina Rigo, Gyorgy Horvath, Tim Hague. Omixon Biocomputing Ltd, Budapest, Hungary. Aim: Sanger nucleotide traces are used in many laboratories and clinics to get accurate and high resolution HLA types, and Next-Generation Sequencing (NGS) reads are just starting to be used for this purpose. Having both capillary and NGS reads from HLA genes makes it possible to study these different approaches to HLA Typing. We present two different NGS HLA typing analysis methods and compare their results to capillary trace based typing. Methods: Eight human samples, with amplicons covering HLA-A,B,C and HLA-DRB1 were analyzed. These amplicons were amplified and sequenced by Sanger capillary sequencing, Illumina MiSeq and IonTorrent instruments. The resulting NGS reads were typing by two different HLA typing algorithms. The first algorithm is a probability-based approach which distributes reads among allele candidates and selects the most likely ones using a comparison of coverage statistics. The second algorithm is a consensus-based approach that aims to deliver phased, positioned consensus sequences and then associate these with the alleles that are supported by the most reads from the provided sample data. Results: Capillary reads can be ambiguous, and covering long genomic parts with Sanger traces can be difficult and impractical. However, Sanger traces are the longest available ones with high quality. Statistics based typing can produce reliable results, but can lose local information (i.e. important SNPs). In addition, phase ambiguities, reads from homologous regions and PCR cross-overs may render statistical typing less reliable. Consensus based typing is less sensitive to homologous or PCR cross-over reads, and can deliver long consensus sequence stretches. This makes possible to discover new alleles or extend existing incomplete ones. Conclusion: NGS brings the hope of covering whole genes, but to assign reads among homologous genes and similar alleles has its own challenges. Comparing the three typing methods reveals differences in dealing with phase ambiguity, novel allele handling and quality control. NGS makes it possible to fully resolve phase and obtain long consensus sequences.