#behavioralgenetics

Dr. Robert Polmin introduces his idea of a Learning Chip that DNA analysis will give each child to serve as a “reliable genetic predictor” of strengths and weaknesses.  Polmin raises an excellent point when he says, “we wait for problems such as reading disability to develop.  Children go to school, they fail, they get diagnoses, they’re given special resources but by then it is too late.”  A lot of the time the very thing is deemed to be true. By DNA analysis giving these children a Learning Chip, Polmin believes that schools will better be able to personalize to each individuals learning and therefore offer a very wide range of choice subjects and extracurricular activities.  Many people might believe that this “learning chip” could predict the difficulties once the person has the gene, and ultimately prevent them.  Other people might believe that such a chip is not physically possible at predicting these difficulties and preventing them.  There are several concepts that support his idea, but there are also several that reject his idea.

 This Learning Chip has the potential to be successful in determining genes and sequencing for specific traits.  By doing this, individualized education would be possible.  Gene environment correlation (rGE) says that genetic propensities are correlated with individual differences in experience and that is adds to phenotypic variance for a trait.  Evocative experiences are a large part of the gene environment correlation, which says individuals, on the basis of their genetic propensities, evoke reactions from other people.  By giving a child a learning chip and being able to predict the strengths and weaknesses of specific genes, teachers can pick out individuals in school and give them special opportunities that best fit them.

 This Learning Chip would essentially run a genome-wide association study (GWAS), which is an examination of many common genetic variants in different individuals to see if any variant is associated with a trait.  This specifically focuses on associations between SNPs and traits like major diseases.  By scanning the whole genome, we can genotype 1-2 million SNPs.  The GWAS is not impossible, but it is certainly not easy.  A very large sample size is required because most traits are influenced by many more genes than we initially thought.  The results of the Learning Chip running the GWAS are that most common diseases have highly polygenetic architecture, genetic effect sizes of common SNVs are very small, genes and biological processes not previously suspected as being involved have been identified, and some loci are involved in several different diseases once thought to be completely independent in terms of etiology.  GWAS could allow for the Learning Chip to determine if a specific gene contributes to a specific trait in children.

 This chip may also run a genome wide complex trait analysis (GCTA).  This allows you to estimate h2 from measured genes.  It also provides an independent measure of heritability, and evidence that a lot of the SNPs we are measuring are contributing to the trait.  This may seem like it can predict difficulties in children, and maybe it can, but the GCTA usually explains only about 50% of h2 estimated from family studies and does not take into account the familial genes.   The other 50% from common SNPs is missing from the GCTA, which is referred to the missing heritability problem.  Even with larger and larger sample sizes, when you add up the significant GWAS SNPs, it still only accounts for a small portion of the heritability.  This information is missing for several reasons including, there are more variants than we thought, there are rare variants, structural variants are present, etc.  With this much information missing it would be hard for the Learning Chip to accurately predict genetic difficulties.

 Rare variants are a mutation on a specific gene.  A rare variant is a gene that is so rare that it has an allele frequency of that is less than 1% of the population.  This would certainly raise some controversy over the Learning Chip and how reliable and accurate it is at predicting difficulties.  If the chip scans a rare variant in an individual, that individual could be placed in a class that is not best suited for him or her.  Because rare variants may disrupt the effectiveness of the Learning Chip, highly polygenetic traits must be considered.  These highly polygenetic traits are not caused by a single gene, but rather multiple genes.  They provide an independent measure of heritability that is not subject to problems with family studies, and it also provides evidence that a lot of the SNPs that we are actually measuring are contributing to the specific trait. 

 Another issue that the Learning Chip would come across that would make it implausible would be gene environment interaction (GxE).  This is genetic sensitivity or susceptibility to the environment, and the effects of the environment depends on the genotype, and vise versa.  To put it short, peoples genes affect what environments they experience, and all people experience the same environments, but particular combinations of genes and environments have different effects.  The development of children has a large part to do with the environment, and the Learning Chip would not account for that.