I started a fellowship at Harvard this fall, and it’s taken me a while to find the time to sit down to do a round-up. There’s been a torrent of news, covering DTC player expansions, major new initiatives, and law enforcement’s use of DNA technology – as well as a host of other topics. We’re a year out from the birth of the first genetically modified babies, and full swing into the polygenic turn. Diversity remains a governing watchword.

Controversy

• Following on from the genetics of non-heterosexuality paper, GenePlaza added an App based on the genetics of non-heterosexuality paper published earlier this year. Ben Neale, lead author of the paper, wrote to GenePlaza urging them to take it down, saying that the paper explicitly stated that individual level prediction was impossible.The App is no longer available. A whole set of other Apps based on polygenic scores continue to be available. Scientists who claim that individual usage in this context is outright wrong are involved in individual usage in clinical applications. This is an ethical line, not a scientific one.
• Researchers who helped gather African genotype data in collaboration with the Sanger have demanded the samples back. They heard that there was a commercial tool planned, and say that this wasn’t within scope of the data sharing.
• One year on from the CRISPR babies
  • He Jiankui’s paper about Lulu and Nana was turned down from two journals and from BioArxiv for ethical reasons. Now MIT Tech Review has published excerpts, alongside commentary.
  • An ethicist makes the case for germline modification, on utilitarian grounds.
  • Russian authorities have called gene edited babies premature, a long awaited announcement from a country with no laws banning the procedure, and with a scientist who has indicated he would go ahead with a green light.

Science

• New initiatives
  • The 500,000 genomes of UK BioBank participants will be sequenced, at a cost of £200m, £100m of which will come from pharma companies. Target completion date is Summer 2021.
  • The UK is working on sequencing 5 million genomes. And Matt Hancock, the Health Secretary, wants to do newborn sequencing ASAP, “Predictive, preventative, personalised healthcare – that is the future of the NHS – and whole genome sequencing and genomics is going to play a huge part in that.”
  • The International Common DIsease Alliance launched, aiming to tackle “maps to mechanisms to medicines”. Eric Lander describes this as the next major push of the genomics community.
  • The NIH is funding, to a tune of about $30m, the construction of a new “pangenome” reference genome, to be based on 350 individuals of diverse ancestries.
  • The NIH is planning a Polygenic Risk Score Consortium, with $34m available (here and here). 
• A study of 426 people from 13 African countries shows just how diverse genetic ancestry is in Africa, and hence how inadequate just incorporating one ethnic group from Africa is in studies.
• Lupus disproportionately impacts black women. Epigenetic differences have now been tied to this difference, “highlighting the lack of information scientists have regarding racial differences on the genetic level”. The familiar refrain about more diversity in samples follows.
• Genetic variation that leads to less need for sleep. Two members of a family with this variant need much less sleep – 5.5 and 4.3 hours – than average. This reminds me of the plot of Nancy Kress’s Beggars from Spain, which imagines children genetically engineered to need no sleep. 
• Results of levels of trust in data sharing amongst the public are out from the Your DNA Your Say project (N ~9000, across USA, UK, Canada, Australia). 19% would trust the government of their country.
• A write-up of a study that purports to use AI to find three gene and environmental networks that shape personality.
• An antisense oligonucleotide therapy developed for one girl seems to have helped her. This is true personalized medicine. In an editorial, FDA members, ask “In these “N-of-one” situations, what type of evidence is needed before exposing a human to a new drug? Even in rapidly progressing, fatal illnesses, precipitating severe complications or death is not acceptable, so what is the minimum assurance of safety that is needed? How persuasive should the mechanistic or functional data be? How should the dose and regimen be selected? How much characterization of the product should be undertaken? How should the urgency of the patient’s situation or the number of people who could ultimately be treated affect the decision-making process? In addition, how will efficacy be evaluated?” They hint at something that I think has to be the way forward: “Approvals as variations on a well-characterized archetypal product might be feasible if the interventions are closely related.” 
• A preprint pointing to the genetic correlates of general intelligence, g. Various different cognitive measures are correlated; so is the underlying genetics.
• A large scale study looking at the genetics of schizophrenia in European and East Asian populations. To quote their major conclusions:
  • “When a single population is used to identify the disease-associated loci, the discovery is skewed towards disease-associated variants that have greater allele frequency in that population. When multiple populations are used, disease-associated variants are equally represented across the allele frequency spectrum in these populations. This shows that including global samples improves the power to find disease associations for which the power varies across populations.”
  • “For traits such as body mass index and autoimmune diseases, we observed heterogeneity across populations in genetic effects, which may point to interactions between genetic associations and environmental factors and/or other genetic loci. In contrast, for schizophrenia, we did not find significant heterogeneity across EAS and EUR ancestries.”
  • “Despite a cross-population common variant genetic correlation being highly consistent, we found that polygenic risk models trained in one population have reduced performance in the other population due to different allele frequency distributions and LD structures. This highlights the importance of including all major ancestral groups in genomic studies, both as a strategy to improve the power to find disease associations, and to ensure that the findings have maximum relevance for all populations.”
• In a perspective piece, Reading tea leaves? Polygenic scores and differences in traits among groups, Graham Coop explains “the ways in which issues of causality, stratification, gene-by-environment interactions, and divergence among groups all complicate the interpretation of among-population polygenic score differences”
• The BabySeq project returns adult onset information to families of newborns. But ethicists continue to argue that this risks a child’s right to an open future.
• CRISPR could be a major weapon in the antibiotic arsenal.
• Can Biology Class Reduce Racism? A class designed to communicate the complications of genetic science to high school students shows promise.

Applications

• Ancestry is going to offer health information to its customers. The tests will be ordered through a physician (who will not interact with the patient), and will include genetic counseling. The focus will be on 17 genes. Customers will be able to a SNP Chip based add on to their ancestry results for about $49, or opt in to NGS exome testing, which has an initial $199 price tag and a $49 per six months for them to reinterpret data. The initial focus of this will be on the same 17 genes. The subscription model is a bold attempt to address the issue that the underlying evidence base is changing. No doubt they’ll also be excited that the once every six months will increase customer engagement too.
• 23andMe has partnered with TrialSpark for recruitment to clinical trials. Potential research participants would be identified based on their phenotype and/or genotypic information. 23andMe can capitalize on its very engaged user base: “The company estimated that currently, more than 80 percent of its customers have opted to participate in research and over 60 percent have logged in to their account in the last 90 days.”
• Anna Wojcicki, founder and CEO of 23andMe, thinks that we just need time for people to get used to the idea of sharing genetic information, just like we needed time to get used to putting our credit card information online. “The reality is with a new technology, it just takes time for people to be comfortable with it.”
• SNPedia and Promethease have been bought by Israeli based myHeritage
• Nebula Genomics is offering anonymized sequencing. Enabled by blockchain.
• Illumina and the Broad are collaborating on a GATK based variant calling platform. The results will be open source, but Illumina will also develop a proprietary version that will run on its hardware Dragen (formerly of Edico).
• Wired reports on the progress made by 54gene. Operating in Nigeria, the VC backed company is getting access to samples and medical records through additional consent given by participants of other clinical trials. They’re on track for 200,000 samples by the end of 2020.
• I’ve written about Genomic Prediction, the company that will test embryos for a range of polygenic scores. They report their first pregnancy post embryo polygenic profiling. This is a long time after their test has been on the market.
• Genomics Medicine Ireland  is under increasing scrutiny – it aims to sequence 10% of the country’s population, and unlike other major national initiatives, it is a private enterprise.
• China is using tech for predicting faces from DNA on the Uighur population.
• Performance enhancing probiotics tailored to an athlete’s microbiome? There’s a startup out of Harvard working on it. 

Regulation etc

• DNA and law enforcement
  • The LA times reports on a backlash against the use of rapid DNA analyzers as part of booking, mostly because of a rushed roll out, rather than in principal concerns. “Roughly 30 states allow immediate DNA testing of crime suspects — including California, which permits it for all felony arrestees. Several states have no such laws, or limit such testing to only the most serious felony cases.” 
  • Department of Justice proposals to collect DNA from everyone who crosses the border illegally. 
  • An interim Department of Justice policy restricts the use of genetic genealogy to a) cases where there has not been a match in CODIS, and b) violent crimes or unidentified remains.
  • GEDMatch, the database that individuals can upload their own data to, gave its users the choice whether to let their data be accessible to law enforcement. Only 185,000 of over 1 million did so. Now, a Florida detective was able to secure a warrant to search the whole database. The concern is that this will set a dangerous precedent, giving law enforcement confidence to go after 23andMe and Ancestry.com.
• Someone serving a sentence for murder is being allowed to appeal on the basis of carrying the allele of MAOA associated wuth heightened aggression.
• A right to know? Or not? In the UK, a woman is suing because she was not told that her father had tested positive for Huntingtons, based on his wishes. In Germany, a woman is suing because she was told that her ex-husband tested positive.
• New regulations will make it easier for Apps to get your health data. But once they have it, it will not be subject to much by way of privacy protection, as the Apps will not be covered entities under HIPAA.
• The Regulatory and Ethics Work Stream (REWS) of the Global Alliance for Genomics and Health (GA4GH) has updated its policies, which touch on all aspects of data sharing. The aim, as explained by group chair Prof. Bartha Knoppers, is to “provide guidance that will support the human right to benefit from this work”
• GDPR requires “adequacy” determinations with other countries before data can flow freely. The first such one has been granted, to Japan. The decision reflected a willingness on the part of the EU to be flexible, in instituting supplemental rules.
• In France, all DTC genetic tests are banned, with criminal penalties. This includes genealogical tests. Some want to keep it that way for now, because if the ban is lifted, foreign companies will rush in.