Tuesday, May 29, 2012

Random enrichment of minor alleles of common SNPs affects complex traits and diseases

We have recently submitted a manuscript for publication as well as an abstract to a human genome variation meeting.  The work represents an application of the MGD hypothesis in solving major real world biomedical problems.  The results confirm the MGD hypothesis and invalidate the neutral theory that has prevailed for nearly half of a century.  The flawed neutral theory has no real world relevance to major biomed problems and no neutral theory experts are known to have contributed anything to research of major biomedical importance (not Kimura, not Ayala, not Nei, not Blair, not Avise, not Felsenstein).  In contrast, the MGD hypothesis can and will survive even without relying on its value to evolution research.  Much of our ongoing research have nothing to do with evolution.  Stay tuned for more of our results on the genetic basis of complex traits and complex diseases about which the neutral paradigm has absolutely no clues. 

Random enrichment of minor alleles of common SNPs affects complex traits and diseases

Dejian Yuan1#, and Zuobin Zhu1#, Xiaohua Tan1, Jie Liang1, Ceng Zeng1, Jiegen Zhang2, Jun Chen2, Long Ma1, Ayca Dogan3, Gudrun Brockmann3, Oliver Goldmann4, Eva Medina4, Xian Man1, Ke Yi1, Yanke Li1, Qing Lu1, Yimin Huang1, Dapeng Wang5, Jun Yu5, Hui Guo1, Kun Xia1, and Shi Huang1*

1State Key Laboratory of Medical Genetics, Central South University, 110 Xiangya Road, Changsha, Hunan 410078, China; 2High Performance Computing Center, Modern Educational Technology Center, New Campus, Central South University, Changsha, Hunan 410083, China. 3Department of Crop and Animal Sciences, Faculty of Agriculture and Horticulture, Humboldt-Universität zu Berlin, Invalidenstraße 42, 10115 Berlin, Germany4Infection Immunology Group, HZI – Helmholtz Centre for Infection Research, Inhoffenstraße 7, 38124 Braunschweig, Germany5CAS Key Laboratory of Genome Sciences and Information, Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing 100029, PR China; Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China. 
*Correspondence to:  Shi Huang, huangshi@sklmg.edu.cn  
#These authors contributed equally to this work.

The prevailing null hypothesis in population genetics and molecular evolution posits that most common SNPs are neutral but this has yet to be formally tested by experimental science.  We employed two strategies to determine whether the minor alleles (MAs) of common SNPs are minor because of natural selection.  First, we analyzed multiple panels of genetic reference populations or recombinant inbred lines (RILs) in model organisms (yeast, worm, fly, mouse, and rat), and identified the MAs of common SNPs in each panel and the fraction of MAs that each strain carries.  We measured the brood size of 104 C. elegans RILs and did genotype-phenotype correlation analysis for the brood trait as well as for ~4700 published and unpublished traits for various RIL panels archived at GeneNetwork.  Although beneficial to immunity as is expected, more MAs correlated significantly with poor measurements in many adaptive traits, including reproductive fitness, life span, tumor susceptibility, anxiety, depression, startle response, and learning and memory.   In addition, more MAs were significantly linked with sensitivity to alcohol, methamphetamine, cocaine, pain, and antipsychotic drugs, and levels in glucose, resistin, insulin, IL-17, iron, and dopamine.  Random enrichment of MAs of common SNPs accounted for as much as 49% of total trait variation in some phenotypes in RILs such as transferrin saturation and food intake.  The majority of the ~4700 traits examined did not significantly correlate with MAs, including blood pressure and morphine response.  Different MA-linked traits may or may not share the same set of MAs with related traits sharing more MAs than non-related traits.  Second, we analyzed 21 published GWAS datasets of common diseases and identified the MAs of common SNPs in each control population and the fraction of MAs each control or case carries.  In Europeans or European Americans, more MAs were significantly and repeatedly linked with type 2 diabetes, Parkinson’s disease, psychiatric disorders, autoimmune diseases, alcohol and cocaine addictions, lung cancer, less life span, and lower education level achieved, but not hypertension and opiate and marijuana addictions.  Thus, the effects of excess MAs in humans are remarkably similar to those in model organisms, suggesting that the link between excess MAs and diseases/traits in humans is causal since it can be replicated in model organisms.  These data indicate that most SNPs in any species are functional or under Darwinian natural selection and open a new avenue of inquiry into the genetic basis of complex traits/diseases.  They confirm a self-evident intuition in construction that any system of great order can allow some random errors/noises in building blocks but only to a limit. 


Grant Dexter said...

What is "MA" and "RIL"?

gnomon said...

MA: minor allele of a SNP. A SNP typically has just two alleles and the allele with a population frequency of <0.5 is termed the minor allele.

RIL: Recombinant inbred lines. The RIL panels are derived from breeding of parental strains differing in phenotypes and genotypes. The F1 and F2 or up to F10 progenies are intercrossed to maximize random recombination and hence allelic diversity in the offspring, which were then randomly selected for inbreeding up to 20 generations to generate the final panel of RILs homozygous for almost all variants or SNPs. During the random mating and subsequent inbreeding process, there are ample opportunities for harmful variants to be negatively selected and for neutral variants to drift. Since natural selection already occurs at fertilization, many individuals may die or be aborted before birth due to abnormal development caused by harmful variants. So, while the population size of a RIL panel is small, which favors neutral drift, the actual size of the offspring population of the original parents is much greater and includes many that were never born due to negative selection of harmful variants. Therefore the frequencies of variants that exist in the established RIL panel are the result of natural selection and/or neutral drift.

Grant Dexter said...

Thanks. :)