Table 3.
Haplotyping software for hypothesis testing and analysis.
| Program name | Haplotyping algorithm | Key analysis feature(s) | Discrete outcomea | Continuous outcomeb |
|---|---|---|---|---|
| Hypothesis | ||||
| CHAPLIN | ECM | Includes likelihood ratio statistic and score statistic for haplotype-phenotype analysis, uses permutation test to determine significance | Yes, case-control | No |
| Includes AIC for model selection, does not accommodate covariates | ||||
| EH | EM | Test for LD for unrelated and in case-control | Yes, case-control | No |
| Test for frequency difference between case-control under: H1 association, H2 association for all loci | ||||
| EHPLUS | EM | Improves on EH | Yes, case-control | No |
| Model-free analysis and permutation test | ||||
| FASTEHPLUS | EM | Implements EH and EHPLUS test | Yes, case-control | No |
| Significant speed improvements | ||||
| GENECOUNTING | EM | Compares overall and specific haplotype frequency between cases and controls | Yes, case-control | No |
| HAPH | IP | Phylogeny based haplotyping method | Yes, case-control | Yes |
| Uses information from phylogeny for analysis, includes parametric and non-parametric tests for qualitative and quantitative phenotypes | ||||
| HAPLO.STATS | EM | Score statistic for haplotype-phenotype analysis | Yes, binary, ordinal, & Poisson | Yes |
| GLM for regression of trait on haplotype, adjustment for covariates and interaction | ||||
| HPLUS | EE +PL+ EM | Compares haplotypes frequency between cases and controls, option to adjust for covariates, and interaction assessment | Yes, case-control | No |
| Reports OR, confidence interval and identifies haplotype blocks | ||||
| LDSUPPORT | EM | Uses likelihood method to calculate risk of developing disease phenotype from diplotype configuration | Yes, case-control | No |
| PHASE v2.0 | MCMC | Allows comparison of haplotype frequency between populations | Yes, case-control | No |
| THESIAS | S-EM | Compares haplotypes frequency between cases and controls, survival analysis, option to adjust for covariates and interaction assessment |
Yes, case-control, survival analysis |
Yes |
| Uses Chi-square statistics/t-test for analysis | ||||
| SAS Genetics | EM | Allows comparison of haplotype frequency between populations | Yes, case-control | Yes |
| Haplotype trend regression (HTR) and several population genetic tests | ||||
| TDT test for family data | ||||
| SNPEM | EM | Compares overall and specific haplotype frequency between cases and controls | Yes, case-control | No |
| Includes batch feature for sliding windows analysis | ||||
| WHAP | EM | Uses SNPHAP for regression based haplotype association test on SNPs, provides beta estimates of effects | Yes, case-control | Yes |
| Includes haplotype weighted likelihood analysis, permutation tests and sliding windows analysis |
||||
| Zaykin et al. | EM | Likelihood ratio statistic for haplotype-phenotype analysis | Yes, case-control | Yes |
| Allows sliding windows analysis | ||||
| 3locus.PAS | EM | Test for global disequilibrium, including pairwise and three | No | No |
| way disequilibrium for an unrelated sample | ||||
| Other analysis programs | ||||
| Arlequin v2.0/3.0 | EM/ELB | Several population genetic tests | ||
| Zou and Zhao | EM | Adjust haplotype frequency estimates for genotyping error |
a Qualitative phenotype.
b Quantitative phenotype.
AIC: Akaike's Information Criterion; EE: Estimating equation; ELB: Excoffier-Laval-Balding algorithm, Bayesian; ECM: Expectation Conditional maximisation algorithm; EM: Expectation maximisation algorithm; GLM: General Linear Model; IP: Imperfect phylogeny; MCMC: Markov Chain Monte Carlo algorithm, Bayesian algorithm; OR: Odds Ratio; PL: Partition ligation algorithm; S-EM: Stochastic EM algorithm.