Genetics, Intelligence, Social Mobility and Chinese Whispers

This week saw the new Government in the UK swing into action. The promises in the Conservative manifesto are coming home to roost and there are many challenges to face as the UK crawls back from the austerity years. In terms of education, the idea of ‘levelling up’ has taken hold. But there are niggling doubts about what this might mean to those in power and how it might be achieved. Further concerns are being raised about the role of genetic determinism in intelligence and attainment for individuals. Those advising in No10 are considering a role for genetic testing of individuals based upon the current interpretation of a vast quantity of DNA sequence data. But science doesn’t deal in absolute ‘truth’. It deals in hypotheses and models that describe how nature works by using the best data available. Misunderstanding of science and how it works is as old as the hills and the dangers of misplaced policies, crafted by politicians who are not scientists, are lurking all the time. 

What is the Government thinking?

This week, former Education Secretary, Justine Greening issued a hopeful statement about social mobility. Having stepped down from government, she is pursuing the aims of her campaign ‘Social Mobility Pledge’ (see TEFS 6th September 2019 ‘Justine Greening…Greening… Gone!’). This week she also called upon the Government to improve social mobility as proposed in her policy paper of December 2017 ‘Improving social mobility through education’. Its main premise was that “While talent is spread evenly across our country, opportunity is not. None of us should accept this.” The aim is to rectify this and promote “Levelling Up Opportunity” across all education. Boris Johnson adopted the slogan ‘levelling up’ in his 2019 conference speech and it was deployed liberally throughout the election campaign.

However, there are many suspicions that the notion of ‘levelling up’, and how to achieve it, means something different at the heart of No10. TEFS earlier reported that Johnson is in favour of envy and greed in the economy (TEFS 9th December 2019 ‘It’s all about equality, Brexit, the environment and the economy, not envy and greed’). In his speech at the Centre for Policy Studies, he also refers to equality with, “it is surely relevant to a conversation about equality that as many as 16 per cent of our species have an IQ below 85, while about 2 per cent have an IQ above 130”. Then, to rub it in, he refers to “the natural and God-given talent of boardroom inhabitants” to support the idea that in the economy “some measure of inequality is essential for the spirit of envy”. This seems counter to ‘levelling up’ as an aim. 

Advice and influence are important.

By appointing Dominic Cummings as his main adviser, Johnson seems to have cemented these ideas into the foundations of policy. Cumming’s proposals to recruit new talent to Government have attracted considerable media flak (see TEFS 3rd January 2020 ‘Social Mobility positions available: Only “Super-talented weirdos” need apply’). But looking deeper into his beliefs causes alarm bells to ring. In his Essay on an ‘Odyssean’ Education from 2013 he defends criticism of a much longer treatise (‘Some thoughts on education and political priorities’) and plunges genetics, ability and social mobility into one boiling pot. He concludes that “Most of those who now dominate discussions of issues such as ‘social mobility’ entirely ignore genetics and therefore their arguments are at best misleading and often worthless.” In February last year, he posted a blog ‘Genetics, genomics, predictions & ‘the Gretzky game’ — a chance for Britain to help the world’. He concluded that a person’s genetic makeup is the main driver of success and most interventions would not work; such as providing disadvantaged children with books. He dismisses the argument that “Kids who can read well come from homes with lots of books so let’s give families with kids struggling to read more books” and replaces it with the “truth” that “children and parents share genes that make them good at and enjoy reading, so causation is operating completely differently to the assumptions”.

The result is a call for “free universal ‘SNP’ (single nucleotide polymorphism) genetic sequencing” for everyone and a polygenic score can then be calculated for various traits including intelligence. The logical outcome would be the use of genetic testing to determine what individuals might receive in terms of health care, social services and education. But whilst such testing might inform the probability of a person in a population, for example, developing a disease in time, it does not predict that an individual will develop the disease. Similarly, it might suggest the probability of an IQ test result but it doesn’t mean that this is the actual IQ of any individual. The technology might be useful in research into populations, and in recognising what genetic factors increase the probability of disease, but it is still imprecise. Furthermore, not all known genetic markers or SNPs are covered in current testing and there are many more confounding influences to consider.

But where is Cummings (a history graduate from Oxford) getting this from and is he convincing Johnson (a Classics graduate from Oxford)? It seems that a popular text ‘Blueprint: How DNA makes us how we are’ by behavioural psychologist Robert Plomin is the source. Cummings says that “I would encourage journalists who want to understand this subject to speak to Plomin who works in London and is able to explain complex technical subjects to very confused arts graduates like me”. The dangers of misleading those in power, whether intentional or not, should not be underestimated. Passing advice on could end up like the ‘telephone message’ or ‘Chinese whispers’ party game. The phrase “genetics determines intelligence sometimes. The environment is more of an influence” might become very quickly “Genetics determines intelligence. Sometimes the environment has no influence”.

The ‘social health warning’ at the start should be on every such communication regarding science being translated into policy.

The dangers of misrepresenting genetics.

The misrepresentation of science to those in power has also been around for a long time. Prominent 20th Century geneticist, Theodosius Dobzhansky (1900-1975) worked at Columbia University and contributed greatly to the understanding of genetics in the middle of the last century. He moved to the USA from Ukraine at the age of 27 and had a distinguished career. His life as a scientist was contemporaneous with fellow Ukrainian Trofim Lysenko (1898-1976) who worked on crop biology and genetics in the Soviet Union. However, their scientific pathways diverged to a huge extent. If Dobzhansky had stayed in the Soviet Union, he would have most likely been arrested for his contrary views and contribution. In the days before the genetic code was known, Lysenko was convinced that by training wheat to yield winter crops out of season, the trait could be also inherited in subsequent crops. This flew in the face of the accepted rules from Gregor Mendel’s analysis of how genes are stably inherited. But the real problem was that Lysenko also convinced others, including a powerful government, that huge crop gains could be made. The forced collectivisation of agriculture had already led to tragic famine and adoption of Lysenko’s ideas as a remedy only exacerbated the situation. The lesson of misrepresenting science to those in power should serve as a warning to all scientists.

The idea of genes controlling our ‘intelligence’ has been around for a very long time.

Dobzhansky wrote a highly acclaimed genetics text in 1937, ‘Genetics and the Origin of Species’, that consolidated thinking on the subject at that time. His observations and insight into the basis of variation in populations and the role of random events in evolution still hold today. In 1973, his essay ‘Nothing in Biology Makes Sense Except in the Light of Evolution’ in ‘The American Biology Teacher’ journal eloquently countered the argument for a creationist view of life and imposed the stricture that we should always consider our observations in the context of evolution and the mechanisms that drive it.

Dobzhansky was also not shy in commenting upon the social implications of science. He published a key paper ‘Genetics and Equality’ in the Journal Science in July 1962. The sexist language prevalent at the time is all too evident and describing the members of ‘Mankind’ exclusively as ‘he’ and ‘him’ grates in our more enlightened times. However, if you are able to press on past this, and accept that Dobzhansky was all too aware of the genetic basis of sex, then his fundamental arguments still have relevance today; despite the expansion in human gene sequencing since. The paper was published a few weeks before Crick, Watson and Wilkins were awarded the Nobel Prize for their discovery of DNA structure, based upon the experimental work of the late Rosalind Franklin back in 1953. By 1962, the structure of DNA was embedded in the subject of genetics. The so-called ‘central dogma’ of DNA coding for RNA and then protein sequences had been postulated. This confirmed the causal link between DNA sequence and the functions of specific proteins that governed cellular functions and development. The link between sequence and function was evident to Dobzhansky. In turn, he foresaw the ethical dilemmas and issues that persist today.

It is worth revisiting these arguments and their implications since they have not receded in time. He starts with the assertion that “All men (sorry there was a warning) may have been created equal; most certainly they are not all alike.” He goes onto explain why “equality” and “identity” are confused both “chronically” and “persistently” alongside “diversity” and “inequality”. Emphasis is put on the observation that “no two persons living have the same genotype” and that “What we inherit is, however, not fixed qualities, but potentialities”. In assessing the role of genetics in the “inheritance of behaviour” and “social mobility”, he concludes that there is likely to be a wide degree of variation. He favours the “adaptive norm of a species or population as a great array of diverse genotypes, heterozygous for many genes, adaptive to different environments and ways of life.” Recent advances in genetics confirm this approach was correct. The large numbers of single nucleotide polymorphisms (SNPs) or variations in gene sequence feed the idea of the cumulative effect of multiple genes working together to influence behaviour. This idea dominates our current thinking. But the Dobzhansky caveat that “restriction of social mobility frustrates the genetic selection” might be considered more today.

It is interesting to note that around about the same time Robbins was working hard on the Higher Education report that emerged a year later (Robbins Committee on Higher Education Report, 1963). It is likely that the ideas of Dobzhansky had crossed the Atlantic to the UK and influenced education policy. They partly explain why the committee Robbins chaired became determined to open the universities more widely so that places should be "available to all who were qualified for them by ability and attainment". This became known as the 'Robbins Principle'. TEFs highlighted the report’s allusions to the role of genetics and the “pool of ability” in ‘Augar and the dark side of Robbins’ (TEFS 7th June 2019). The expansion of the higher education system post-Robbins was based upon the assertion that there was a very wide pool of ability. This was considered much greater than the then HE provision could cope with and huge investment was needed.

Popular science today and the possibility of misleading the public.

The hidden danger of communicating science to the public in simpler terms is that incorrect inferences and conclusions could be made. Indeed, writing this post could fall into the same trap and could be prone to causing misunderstandings. This would be an untended consequence and may have little adverse effect for the most part. But a real danger lies in those with authority and power formulating policy built around a misunderstanding. All such communications should have a ‘social health warning’ attached such as that at the top of this post. Most of our politicians are not scientists and they may hold naïve views on science. The divide between the ‘two cultures’ in our education system persists. Originally identified by Charles (C P) Snow in 1959 in a Rede Lecture it grew into an influential eponymous book (see TEFS 28th September 2018 ‘Labour Party Conference 2018: National Education Service and a tale of Two Cultures’). One fundamental mistake might be in concluding that scientific ‘fact’ implies ‘truth’. Science instead uses the best evidence available to construct theories and models of how nature works. It does not deal with 'truth’ as such.

In the area of humans and genes, there are several examples of books that influence opinion. At one end of the spectrum is a book by Alistair Moffat ‘Britain: A Genetic Journey’ from 2017. The text is fascinating and interesting from a historical perspective but a paucity of cited references or evidence to back up the scientific assertions should alert readers. The book was savagely attacked last year by geneticist Debbie Kennett and colleagues in University College London and the University of Melbourne in the Genealogy Journal with ‘The Rise and Fall of BritainsDNA: A Tale of Misleading Claims, Media Manipulation and Threats to Academic Freedom’. The charge that the book and other media reports were also there simply to attract people to the author’s commercial gene profiling company ‘Britain’s DNA’, that is no longer trading, was compelling and attracted threatened legal action. The book also attracted condemnation of the idea that genetic evidence in gene markers from exclusively maternal mitochondrial DNA or exclusively male Y chromosome DNA could determine lineage going back hundreds of years. Whilst it may be possible to infer a relationship between two individuals over a recent time span, inferring longer-term ancestry is fraught with problems. It was established that “the contribution of DNA from any particular ancestor can be nil after just a handful of generations”. The genetic system that drives variation and evolution ensures that this is the case. Yet there is no doubt that people are intrigued and attracted to the idea that they might be related in some way to a famous historical figure.

TEFS suggests that interpretation of sequence data from any individual should be the subject of very strict regulation and oversight.

At the other end of the spectrum is the recent book by Robert Plomin ‘Blueprint: How DNA makes us how we are’. It is influencing Government at the top. As a prominent behavioural scientist, Plomin has been prolific in furthering the evidence that genetic differences determine intelligence and attainment. His book is aimed at the public and is well written. It is also well researched in as far as it goes and cites many sources correctly in notes. Thus, the book is likely to be highly influential among politicians. However, the author avoids some salient issues by omission. A statement at the outset that “some of these tangential topics that I have reluctantly let go include evolution, epigenetics and gene editing”. That we are all the result of evolution is hardly “tangential”, it is central, and the role of epigenetics cannot be avoided. The dangers of deploying gene editing in the future, and thus counteract the mechanisms that ensure diversity, should be fully recognised.

We are all different because of evolutionary processes.

The genetic system and sexual cycle that produces children to maintain the population are very elegant. Complex biochemical mechanisms work together to feed selection and evolution with incredible diversity. The combination of chromosomes XX for female and XY for male ensures that there is a 50% chance of either emerging. Having two copies of each of the other 22 chromosomes and the arrangement of genes ‘packaged’ in chromosomes is not accidental. They are the result of evolution. For evolution to work through selection, there is a necessity for wide variability in the population. The system has therefore evolved to maximise the number of variants possible. Because of this, no two chromosomes have the same sequence at many loci; although they may be generally similar overall. Random mixing of the chromosomes when producing sperm and eggs leads to millions of possibilities. Then random encounters between sperm and eggs magnify this to trillions. The video here depicts the pairs of different chromosomes as letters with copies of the same letter being an upper or lower case to signify that they have differences in sequence or SNPs. 

The number of possible permutations is massive. Imagine your chances of winning the Euromillions Lottery jackpot through guessing one permutation in a sequence of seven from fifty numbers in the pool. Matching 5 numbers and 2 Lucky Stars has a 1 in 139,838,160 chance. Now consider doing the same for all 23 numbers in a pool of 23  with two winners meeting by chance. The outcome is that no two humans will be the same unless they are identical twins arising from the same fertilized egg. It is thus very important to remember that children are not the same in the same family. They may share half of their parent’s genes, but these are mixed in different permutations to give one combination.

Better together.

One analogy might be that of a large orchestra. Listening to players practising the score for a single instrument alone is not very inspiring (in my case the cello when at school – as an aside it was taken from me at sixteen as my parents could not afford to buy one). But together with the other instruments, it is sublime. The emotional power of harmony leaps out at the audience in many of the great classical works.

The conclusion should be that it is the combination of genes together that is more important than the simple possession of shared genes. Just having the instruments for an orchestra is not enough alone. They must be played to elicit an effect. In most cases, the biochemical mechanisms coded by the genes, where there are SNPs, are not known. Thus, any slight variation in function caused by one SNP variant in a gene may not be measurable. A variation in one gene may have no effect when accompanied by a particular combination of SNP variants in other genes. However, when acting ‘in concert’ in the context of another particular combination of genes, the effect becomes evident. With more information about biochemical mechanisms of the brain, we will come to a better appreciation of how the network of genes works together. That time is a long way off.

The role of selection, population mixing and random events.

The current studies designed to identify genes that may be associated together in determining cognitive abilities are essentially ‘snapshots’ in evolutionary time. Observable genetic variation in human populations is shaped by several processes. In addition to ensuring no two individuals are identical, other processes come into play. These include selection, movements of people, populations mixing and their social and economic geography. Many traits are linked together on the same chromosome and this ensures that they are inherited together. The other main driving forces are genetic drift due to random changes in the frequencies of variants in populations and the introduction of mutations over time. The idea that many mutations are selectively ‘neutral’ or ‘near-neutral’ has been largely accepted after the seminal work of Motoo Kimura in the journal ‘Nature’ in 1968 with ‘Evolutionary Rate at the Molecular Level’. An excellent and wider discourse on the issues raised by Kimura is explained by Masatoshi Nei in 2005 in the Journal of Molecular Biology and Evolution with ‘Selectionism and Neutralism in Molecular Evolution’.

Playing from the polygenic score.

The idea behind polygenic scoring (sometimes called polygenic risk score in relation to disease) is that many genes that have very small effects can affect behaviour in a cumulative manner. Adding the effects of many candidate genes leads to a cumulative ‘polygenic score’. This can then be linked to educational attainment or cognitive ability across a population. These studies are genome-wide as the genes and SNPs sequence variants are spread across all chromosomes. The largest genome-wide association study (GWAS) is by James Lee of the University of Minnesota and seventy-seven co-workers. They reported that the number of SNPs associated with educational attainment had expanded to at least 1,271 in a screening of over 1.1 million people. Many of the SNPs were in genes related to brain development and function (in Nature Genetics 2018 ‘Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals’). The work is very complex and not very accessible to most people. It involves advanced statistical methodology, many checks and ‘filtering’ of data from the general ‘noise’ of the millions of SNPs found in genomes. More accessible explanations are in ‘P is for Polygenic Risk Scores’ by Christopher Rayner and ‘A guide to performing Polygenic Risk Score analyses’ by Shing Wan Choi and colleagues. 

A critical overview last September by Graham Coop, Professor of Population Biology and Quantitative Genetics at the University of California at Davis, offers another way to look at GWAS and polygenic scores with 'Reading tea leaves? Polygenic scores and differences in traits among groups.' It is an excellent and accessible explanation of how polygenic scores are added up and what some of the pitfalls may be. It is stressed that the limitation of studies to date to people of Europen origin gives a "European-biased view of genetic variation". The role of population stratification is well illustrated by the hypothetical example of "strong tea drinkers" from England living in Paris causing a spurious correlation to polygenic scores amongst the Parisians.

Much of the ‘noise’ in GWAS studies arises from the accumulation of SNPs over time by the process of genetic drift and neutral effects. Indeed, the whole theory of polygenic scoring is predicated on the SNPs being neutral or near-neutral, but then having a cumulative effect together. Rigorous checks and calibrations are needed to avoid pitfalls and false results. One example is that using individuals from the same family might amplify the importance of some SNPs and bias the results for polygenic scores and educational attainment. Lee and co-workers concluded that “some of the predictive power of the polygenic scores reflects environmental amplification of the environmental effects”. Control for any bias associated with families, parental attainment and “rearing environments that promote educational attainment” is needed. This is an important caveat that might have wider implications.

Environmental influences and epigenetics.

In ‘Blueprint: How DNA makes us how we are’ Plomin shies away from the possible role of epigenetics. Instead, he concentrates on the role of DNA inherited traits in determining cognitive ability and educational attainment. His arguments are compelling and, in concluding that “students in selective and non-selective schools differ in their DNA. Because the traits used to select students are highly heritable”, Plomin questions the wisdom of parents paying for expensive private schools with the assertion “even though schools have little effect on individual differences in school achievement”.

But the epigenetics conundrum adds spice to the notion of heritability by DNA alone. This idea proposes that modifications to the DNA or associated proteins during the lifetime of an individual affects how genes function and may also be transmitted to the next generation. The idea seems to have also disturbed the ghost of Lysenko with molecular evidence. The proposal that genetic sequence is not affected per se, but only how the expression of the genes is regulated, cannot be ignored. There is mounting evidence that DNA is chemically modified by methylation during growth and that this, in turn, alters the expression of genes and their effectiveness. Last year Giacomo Cavalli, of the Institute of Human Genetics at the University of Montpellier and Edith Heard, of the European Molecular Biology Laboratory (EMBL) in Heidelberg, provided an excellent update of the evidence in the journal Nature with 'Advances in epigenetics link genetics to the environment and disease'. An earlier review in 2016 by David Allis, of the  Laboratory of Chromatin Biology and Epigenetics at Rockefeller University and Thomas Jenuwein, of the Max Planck Institute of Immunobiology and Epigenetics in Freiburg, provides a background to how the science developed with 'The molecular hallmarks of epigenetic control'. Both chemical modification of DNA and proteins associated with chromosomes can impact on the control of genes. For those interested in going further into the literature, the commercial Biotechnology company AtdBio from Oxford has an excellent overview of the chemistry involved in 'Epigenetics'.  It is a good primer before proceeding further.

Consider then the observation of increases in general intelligence over the last century, and within one generation, that cannot be explained by inheritance alone. Improvements in education, nutrition and environment might be important factors in overcoming the epigenetic effects. We should not forget that the current data is a ‘snapshot’ of very recent times. James Flynn and colleagues reported in 1987 that there had been recent and widespread gains in IQ with ‘Massive IQ gains in 14 nations: What IQ tests really measure’. The so-called ‘Flynn effect’ has been the subject of much debate and criticism (see ‘A critique of the Flynn Effect: massive IQ gains, methodological artifacts (sic), or both?) yet the underlying data showing the effect refuses to go away.

The hypothesis is that changes in the social and educational environment have improved the situation through epigenetic effects. In one recent example, it seems that the effectiveness of a key receptor in the brain is affected by methylation of the gene involved (see example of ‘Epigenetic variance in dopamine D2 receptor: a marker of IQ malleability?’ By Kaminski et al in 2018) and it can be ‘damped down’ due to stress. This, in turn, appears to affect performance in IQ tests. With an increasing number of candidate neural development genes associated with intelligence in genome-wide analyses (see Lee et al ‘Gene discovery and polygenic prediction from a genome-wide association study of educational attainment in 1.1 million individuals’) it would be prudent to look at methylation patterns more widely. Increasing child poverty is surely stressing the children affected and in turn impacting upon their attainment and development. If the advances in IQ happened quickly within a generation it might be supposed that they could be reversed just as quickly by the same mechanism if social care recedes.

The idea of a threshold effect.

The idea that a ‘threshold level’ of social care and education must be reached before the full potential of children can be realised comes to the fore. Plomin concludes that paying for private education has little effect on “children’s school achievement”. But this is in the context of most children receiving reasonable care at home and education above a minimum threshold. The ‘Law of diminishing return’ may be operating as parents of means try to add more educational advantage with a declining return. With the radical social reforms after 1945, it may be that hundreds of thousands of children previously below the threshold of disadvantage and poverty were lifted-up and allowed to succeed. However, the increase in child poverty since 2010 is therefore very concerning (as reported by the TUC in November 2019 with ‘Child poverty in working households up by 800,000 since 2010’ using government data ‘Statistics on the number and percentage of people living in low-income households in the UK’). Care should be taken by policymakers to at least define where the threshold lies and keep everyone in the warmth and on the right side of it. Reports that the ‘Number of families in temporary accommodation up 64% since 2010’ should have set alarm bells ringing sooner and seem to indicate that the future for those children will not be a good one.

Plomin’s alternate conclusion is that “There may be benefits of grammar and private schools in terms of other outcomes, such as better prospects at university, making connections that lead to job opportunities later in life, and imbuing students with greater confidence and leadership skills”. It is a sad condemnation of our divided and unequal society. He is right in asserting that “Equality of opportunity, income inequality and social mobility are some of the most critical issues in society today”. 

Addressing these issues through selective social support and selective education linked to an individual's polygenic score is a step too far.  But it may become an unintended consequence of taking too selective an interpretation of the science and evidence behind it. Nature does not give up its secrets so readily and there is a long way to go.

Mike Larkin, retired from Queen's University Belfast after 37 years teaching Microbiology, Biochemistry and Genetics.