Genetic engineers at the lab used the new tool to generate a catalog of 71 million possible missense variants, classifying 89% as either benign or pathogenic
Genetic engineers continue to use artificial intelligence (AI) and deep learning to develop research tools that have implications for clinical laboratories. The latest development involves Google’s DeepMind artificial intelligence lab which has created an AI tool that, they say, can predict whether a single-letter substitution in DNA—known as a missense variant (aka, missense mutation)—is likely to cause disease.
The Google engineers used their new model—dubbed AlphaMissense—to generate a catalog of 71 million possible missense variants. They were able to classify 89% as likely to be either benign or pathogenic mutations. That compares with just 0.1% that have been classified using conventional methods, according to the DeepMind engineers.
This is yet another example of how Google is investing to develop solutions for healthcare and medical care. In this case, DeepMind might find genetic sequences that are associated with disease or health conditions. In turn, these genetic sequences could eventually become biomarkers that clinical laboratories could use to help physicians make earlier, more accurate diagnoses and allow faster interventions that improve patient care.
“AI tools that can accurately predict the effect of variants have the power to accelerate research across fields from molecular biology to clinical and statistical genetics,” wrote Google DeepMind engineers Jun Cheng, PhD (left), and Žiga Avsec, PhD (right), in a blog post describing the new tool. Clinical laboratories benefit from the diagnostic biomarkers generated by this type of research. (Photo copyrights: LinkedIn.)
AI’s Effect on Genetic Research
Genetic experiments to identify which mutations cause disease are both costly and time-consuming, Google DeepMind engineers Jun Cheng, PhD, and Žiga Avsec, PhD, wrote in a blog post. However, artificial intelligence sped up that process considerably.
“By using AI predictions, researchers can get a preview of results for thousands of proteins at a time, which can help to prioritize resources and accelerate more complex studies,” they noted.
Of all possible 71 million variants, approximately 6%, or four million, have already been seen in humans, they wrote, noting that the average person carries more than 9,000. Most are benign, “but others are pathogenic and can severely disrupt protein function,” causing diseases such as cystic fibrosis, sickle-cell anemia, and cancer.
“A missense variant is a single letter substitution in DNA that results in a different amino acid within a protein,” Cheng and Avsec wrote in the blog post. “If you think of DNA as a language, switching one letter can change a word and alter the meaning of a sentence altogether. In this case, a substitution changes which amino acid is translated, which can affect the function of a protein.”
In the Google DeepMind study, AlphaMissense predicted that 57% of the 71 million variants are “likely benign,” 32% are “likely pathogenic,” and 11% are “uncertain.”
The AlphaMissense model is adapted from an earlier model called AlphaFold which uses amino acid genetic sequences to predict the structure of proteins.
“AlphaMissense was fed data on DNA from humans and closely related primates to learn which missense mutations are common, and therefore probably benign, and which are rare and potentially harmful,” The Guardian reported. “At the same time, the program familiarized itself with the ‘language’ of proteins by studying millions of protein sequences and learning what a ‘healthy’ protein looks like.”
The model assigned each variant a score between 0 and 1 to rate the likelihood of pathogenicity [the potential for a pathogen to cause disease]. “The continuous score allows users to choose a threshold for classifying variants as pathogenic or benign that matches their accuracy requirements,” Avsec and Cheng wrote in their blog post.
However, they also acknowledged that it doesn’t indicate exactly how the variation causes disease.
The engineers cautioned that the predictions in the catalog are not intended for clinical use. Instead, they “should be interpreted with other sources of evidence.” However, “this work has the potential to improve the diagnosis of rare genetic disorders, and help discover new disease-causing genes,” they noted.
Genomics England Sees a Helpful Tool
BBC noted that AlphaMissense has been tested by Genomics England, which works with the UK’s National Health Service. “The new tool is really bringing a new perspective to the data,” Ellen Thomas, PhD, Genomics England’s Deputy Chief Medical Officer, told the BBC. “It will help clinical scientists make sense of genetic data so that it is useful for patients and for their clinical teams.”
AlphaMissense is “a big step forward,” Ewan Birney, PhD, Deputy Director General of the European Molecular Biology Laboratory (EMBL) told the BBC. “It will help clinical researchers prioritize where to look to find areas that could cause disease.”
Other experts, however, who spoke with MIT Technology Review were less enthusiastic.
Heidi Rehm, PhD, co-director of the Program in Medical and Population Genetics at the Broad Institute, suggested that the DeepMind engineers overstated the certainty of the model’s predictions. She told the publication that she was “disappointed” that they labeled the variants as benign or pathogenic.
“The models are improving, but none are perfect, and they still don’t get you to pathogenic or not,” she said.
“Typically, experts don’t declare a mutation pathogenic until they have real-world data from patients, evidence of inheritance patterns in families, and lab tests—information that’s shared through public websites of variants such as ClinVar,” the MIT article noted.
Is AlphaMissense a Biosecurity Risk?
Although DeepMind has released its catalog of variations, MIT Technology Review notes that the lab isn’t releasing the entire AI model due to what it describes as a “biosecurity risk.”
The concern is that “bad actors” could try using it on non-human species, DeepMind said. But one anonymous expert described the restrictions “as a transparent effort to stop others from quickly deploying the model for their own uses,” the MIT article noted.
And so, genetics research takes a huge step forward thanks to Google DeepMind, artificial intelligence, and deep learning. Clinical laboratories and pathologists may soon have useful new tools that help healthcare provider diagnose diseases. Time will tell. But the developments are certain worth watching.
Millions of cancelled healthcare appointments and lengthy waits for care abound in UK, New Zealand, and in the US
Strikes continue on multiple continents as thousands of healthcare workers walk off the job. Doctors, medical laboratory scientists, nurses, phlebotomists and others around the world have taken to the picket lines complaining about low wages, inadequate staffing, and dangerous working conditions.
In England, junior doctors (the general equivalent of medical interns in the US) continue their uphill battle to have their complaints heard by the UK government. As a result, at hospitals and clinics throughout the United Kingdom, more than one million appointments have been cancelled due to strikes, according to the BBC.
“The true scale of the disruption is likely to be higher—many hospitals reduce bookings on strike days to minimize last-minute cancellations,” the BBC reported. “A total of one million hospital appointments have had to be rescheduled along with more than 60,000 community and mental health appointments since December [2022], when industrial action started in the National Health Service (NHS).”
According to The Standard, “Consultants in England are to be re-balloted over the prospect of further strike action as doctors and the government remain in talks with a view to end the dispute. The British Medical Association (BMA) said that specialist, associate specialist, and specialty (SAS) doctors will also be balloted over potential strike action.”
“We must be prepared to take the next step and ballot for industrial action if we absolutely have to—and we will do this … if upcoming negotiations fail to achieve anything for our profession,” Ujjwala Anand Mohite, DRCPath, FEBPath (above), a histopathologist at the NHS, Dudley Group of Hospitals, and the first female Chair of the SAS committee UK, told The Guardian.
New Zealand Doctors, Clinical Laboratory Workers Strike
In September, the first-ever nationwide senior doctor strike occurred in New Zealand and was then followed by another strike of about 5,000 doctors and 100 dentists from New Zealand’s public hospitals, the World Socialist Web Site reported.
Similar to the UK, the strikes reflect mounting frustration over pay not keeping up with inflation and “decades of deteriorating conditions in the public health system,” the WSWS noted.
This follows months of strikes by the island nation’s medical laboratory workers, which are ongoing.
“Our pay scales, if you compare them internationally, are not competitive. About half of our specialists come from abroad, so it’s quite important for the country’s health system to be able to attract and keep people,” Andy Davies, a lung specialist who joined the picket outside 484-bed Wellington Hospital, told the WSWS.
“We’re not asking for the world, we’re asking for an inflationary pay rise, and we haven’t had an inflationary pay rise year-on-year, and it’s beginning to show,” he added.
“What type of health system do they want?” he continued. “Do we want one that treats all people and manages what they need, or do we want a hacked down system that does less?”
The conflicts over pay and working conditions have caused many healthcare workers in New Zealand to leave the field entirely. This has led to severe shortages of qualified workers.
“Patient waiting times—for cancer, hip replacements, cardiac problems, and many other conditions—have exploded due to understaffed and overwhelmed hospitals,” the WSWS reported.
US Healthcare Workers also Striking
The US has its share of striking healthcare workers as well. Healthcare Dive tracked 23 ongoing or anticipated strikes throughout the nation’s healthcare industry since January 1, 2023. In 2022, there were 15 strikes of healthcare workers at the nation’s hospitals and health systems.
These walkouts include doctors, nurses, pharmacy workers, imaging specialists, and thousands of frontline healthcare workers striking over dangerously low staffing levels, unsafe working conditions, and low pay.
In October, 75,000 nurses, support staff, and medical technicians from Kaiser Permanente participated in a 72-hour strike comprised of hundreds of hospitals and clinics throughout California, Washington state, Oregon, Virginia, and the District of Columbia, Reuters reported.
The three-day strike, “Marked the largest work stoppage to date in the healthcare sector,” Reuters noted. Doctors, managers, and contingency workers were employed to keep hospitals and emergency departments functioning.
“The dispute is focused on workers’ demands for better pay and measures to ease chronic staff shortages and high turnover that union officials say has undermined patient care at Kaiser,” Reuters stated.
Staffing shortages following the COVID-19 pandemic are partly to blame for current struggles, but contract staffing to fill critical positions has exacerbated the problem.
“Kaiser’s outsourcing of healthcare duties to third-party vendors and subcontractors has also emerged as a major sticking point in talks that have dragged on for six months. … The clash has put Kaiser Permanente at the forefront of growing labor unrest in the healthcare industry—and across the US economy—driven by the erosion of workers’ earning power from inflation and pandemic-related disruptions in the workforce,” Reuters noted.
Across the globe, many healthcare workers—including clinical laboratory scientists in countries like New Zealand—are feeling burnt out from working in understaffed departments for inadequate pay. Hopefully, in response to these strikes, governments and healthcare leaders can come to resolutions that bring critical medical specialists back to work.
Research in the UK and US into how rapid WGS can prevent deaths and improve outcomes for kids with rare genetic diseases may lead to more genetic testing based in local clinical laboratories
Genetic scientists with the National Health Service (NHS) in England have embarked on an ambitious plan to offer rapid whole genome sequencing (rWGS) for children and babies with serious illnesses, as part of a larger initiative to embrace genomic medicine in the United Kingdom (UK).
The NHS estimates that the plan will benefit more than 1,000 children and babies each year, including newborns with rare diseases such as cancer, as well as kids placed in intensive care after being admitted to hospitals. Instead of waiting weeks for results from conventional tests, clinicians will be able to administer a simple blood test and get results within days, the NHS said in a press release.
The press release notes that about 75% of rare genetic diseases appear during childhood “and are responsible for almost a third of neonatal intensive care deaths.”
Here in the United States, pathologists and clinical laboratory managers should see this development as a progressive step toward expanding access to genetic tests and whole genome sequencing services. The UK is looking at this service as a nationwide service. By contrast, given the size of the population and geography of the United States, as this line of medical laboratory testing expands in the US, it will probably be centered in select regional centers of excellence.
“This strategy sets out how more people will be empowered to take preventative action following risk-based predictions, receive life-changing diagnoses, and get the support needed to live with genomically-informed diagnoses alongside improved access to cutting-edge precision [medicine] treatments. It also outlines how the NHS will accelerate future high-quality genomic innovation that can be adopted and spread across the country, leading to positive impacts for current and future generations,” the NHS wrote.
“This global first is an incredible moment for the NHS and will be revolutionary in helping us to rapidly diagnose the illnesses of thousands of seriously ill children and babies—saving countless lives in the years to come,” said NHS chief executive Amanda Pritchard (above) in a press release announcing the program. (Photo copyright: Hospital Times.)
New Rapid Whole Genome Sequencing Service
The NHS announced the plan following a series of trials last year. In one trial, a five-day old infant was admitted to a hospital in Cheltenham, Gloucester, with potentially deadly levels of ammonia in his blood. Whole genome sequencing revealed that changes in the CPS1 gene were preventing his body from breaking down nitrogen, which led to the spike in ammonia. He was given life-saving medication in advance of a liver transplant that doctors believed would cure the condition. Without the rapid genetic test, doctors likely would have performed an invasive liver biopsy.
Using a simple blood test, the new newborn genetic screening service in England is expected to benefit more than 1,000 critically ill infants each year, potentially saving their lives. “The rapid whole genome testing service will transform how rare genetic conditions are diagnosed,” explained Emma Baple, PhD, Professor of Genomic Medicine at University of Exeter Medical School and leader of the National Rapid Whole Genome Sequencing Service in the press release. “We know that with prompt and accurate diagnosis, conditions could be cured or better managed with the right clinical care, which would be life-altering—and potentially life-saving—for so many seriously unwell babies and children,” Precision Medicine Institute reported.
According to The Guardian, test results will be available in two to seven days.
Along with the new rWGS testing service, the NHS announced a five-year plan to implement genomic medicine more broadly. The provisions include establishment of an ethics advisory board, more training for NHS personnel, and an expansion of genomic testing within the existing NHS diagnostic infrastructure. The latter could include using NHS Community Diagnostics centers to collect blood samples from family members to test for inherited diseases.
UK’s Longtime Interest in Whole Genome Sequencing
The UK government has long been interested in the potential role of WGS for delivering better outcomes for patients with genetic diseases, The Guardian reported.
In 2013, the government launched the 100,000 Genomes Project to examine the usefulness of the technology. In November 2021, investigators with the project reported the results of a large pilot study in which they analyzed the genomes of 4,660 individuals with rare diseases. The study, published in the New England Journal of Medicine (NEJM) titled, “100,000 Genomes Pilot on Rare-Disease Diagnosis in Health Care—Preliminary Report,” found “a substantial increase in yield of genomic diagnoses made in patients with the use of genome sequencing across a broad spectrum of rare disease.”
The study’s findings suggest that use of WGS “could save the NHS millions of pounds,” The Guardian reported.
Whole Genome Sequencing System for Newborns in the US
“This NBS-rWGS [newborn screening by rapid whole genome sequencing] system is designed to complement the existing newborn screening process and has the potential to eliminate the diagnostic and therapeutic odyssey that many children and parents face,” Kingsmore said in a press release. “Currently, only 35 core genetic disorders are recommended for newborn screening in the United States, but there are more than 7,200 known genetic diseases. Outcomes remain poor for newborns with a genetic disease because of the limited number of recommended screenings. With NBS-rWGS, we can more quickly expand that number and therefore potentially improve outcomes through precision medicine.”
A more recent 2023 study which examined 112 infant deaths at Rady Children’s Hospital found that 40% of the babies had genetic diseases. In seven infants, genetic diseases were identified post-mortem, and in five of them “death might have been avoided had rapid, diagnostic WGS been performed at time of symptom onset or regional intensive care unit admission,” the authors wrote.
“Prior etiologic studies of infant mortality are generally retrospective, based on electronic health record and death certificate review, and without genome information, leading to underdiagnosis of genetic diseases,” said Christina Chambers, PhD, co-author of the study, in a press release. “In fact, prior studies show at least 30% of death certificates have inaccuracies. By implementing broad use of genome sequencing in newborns we might substantially reduce infant mortality.”
Pioneering work with whole genome sequencing for newborns, such as that being conducted by the clinical laboratory and genetic teams at Rady Children’s Hospital and the UK’s NHS, could allow doctors to make timely interventions for our most vulnerable patients.
Project aims to create a new pangenome for genetic testing that will ensure better clinical laboratory testing and healthcare outcomes
Recent advances in genetics are motivating some scientists to proclaim the need to update the existing “master human genome”—currently based on a single individual’s genetic sequence—to make it more inclusive. This international research effort will have implications for personalized clinical laboratory testing and precision medicine.
Genetic scientists at the Human Pangenome Reference Consortium (HPRC), a project funded by the National Human Genome Research Institute (NHGRI), are working “to sequence and assemble genomes from individuals from diverse populations in order to better represent [the] genomic landscape of diverse human populations,” according to the organization’s website.
The project plans to evaluate a wide variety of reference genomes and develop a more diverse human pangenome (a multi-genome reference sequence) that will contain a larger cross-section of the human population. The HPRC scientists will be looking at genomes from specific countries, including Denmark, Japan, South Korea, Sweden, and the United Arab Emirates, The Guardian reported.
The increased diversity of reference genetic data will enable genomic researchers to increase the accuracy of precision medicine diagnostics and clinical laboratory testing.
“One person is not representative of the world,” Pui-Yan Kwok, MD, PhD (above), Henry Bachrach Distinguished Professor, Cardiovascular Research Institute at the University of California, San Francisco, told The Guardian. “As a result, most genome sequencing is fundamentally biased.” And that bias, the researchers claim, affects the accuracy of clinical laboratory treatments and diagnostics. (Photo copyright: UCSF.)
Reference Genome for Genetic Sequencing is Based on One Person
Launched in 1990, The Human Genome Project studied all DNA in a select set of organisms. The project completed its first sequence of the human genome in 2003, which became the reference genome for thousands of genomic discoveries since then.
But there’s a problem.
Although a revolutionary breakthrough in genetic sequencing, that reference genome came from just one person. This means a significant portion of the human population is not represented in genetic research, and that bias, according to some scientists, “limits the kind of genetic variation that can be detected, leaving some patients without diagnoses and potentially without proper treatment,” according to The Guardian.
“Getting the right medicine to the right patient at the right time is the tagline,” Neil Hanchard, MD, DPhil, physician scientist and senior investigator for precision health research at the NHGRI in Bethesda, Maryland, told The Guardian.
The HPRC’s goal is to help mitigate reference biases that could hamper disease diagnoses and ensure all populations receive the best treatments for illness.
According to its website, the organization’s main purpose includes:
Gene sequencing from a diverse set of samples with the newest technologies.
Fostering an ecosystem of assembly and pangenome tools.
Creating and releasing high-quality assemblies and pangenomes.
Embedding a team of scholars to address ethical, legal, and social implications of their work.
Forming international partnerships for the research.
HPRC Scientists Find Never-Sequenced Genetic Variants in Africa
Standard gene sequencing works by dividing DNA into tiny portions known as short reads, then sequencing and organizing the reads into a genome using an existing reference as a guide. However, this process renders larger blocks of variants, called structural variants (SVs), more difficult to read or even remain undetected, which can translate to a sequence that does not completely represent personal variations.
In 2019, the HPRC team of scientists analyzed genetic samples from 154 people from various parts of the world and discovered SV content that was missing from their reference sequence. A further study of genetic samples from 338 individuals that examined only extra inserted DNA detected the presence of almost 130,000 new sequences.
More recently, the HPRC researchers sampled 426 individuals from 50 ethnolinguistic groups from Africa and discovered a few million new single nucleotide variants (SNVs). Most of these distinct SNVs derived from populations that had not been previously sampled.
“We haven’t even touched SVs,” Hanchard told The Guardian. “But our preliminary data suggests it’s going to be more of the same.”
“We may miss risk variants in those regions not represented in the reference,” he added.
HPRC Receives Clearance from NHGRI to Continue Research
Hanchard recognizes the benefits of regional references in genomic sequencing and is optimistic about the future of genomics and the ability to sequence more diverse populations.
“I would love to get to a point where everyone feels represented and that this is for them, as much as it is for any particular group,” he told The Guardian. “We are from one humanity, that’s the important part.”
On February 13, the HPRC received concept clearance for renewal of the program from the NHGRI, which plans to commit up to $10 million in total costs per year for the program over the next five years.
Genetic sequencing continues to emerge as a vital tool in the diagnoses and treatment of diseases. Ensuring that as many diverse populations as possible are included in genomic research is an important element for precision medicine and optimal healthcare.
Clinical laboratory managers and pathologists will want to stay updated on these developments, because much of this new knowledge about the pangenome will need to be incorporated when interpreting genetic sequences and developing diagnoses in support of personalized medicine.
Clinical laboratory scientist who aided in the investigation compared DNA test results with publicly available genetic information
In an interesting twist in the solving of crime, genetic test results—along with help from a clinical laboratory scientist (CLS) turned amateur genealogist—guided relatives of Melissa Highsmith to her whereabouts after she was allegedly kidnapped as a toddler over half a century ago. According to The Guardian, the CLS helped locate Melissa by “interpreting the key DNA results and mining publicly available records.”
Highsmith’s abduction was one of the oldest missing person cases in the country and demonstrates how clinical laboratory skills can be applied outside the laboratory to help solve other problems—in this case, helping a family search for a kidnapped daughter—using genetic testing technologies that until recently were not available to the general public.
Thanks to a 23andMe at-home DNA test—and a tenacious clinical laboratory scientist/amateur genealogist—Melissa Highsmith (shown above at time of kidnapping and today) has been reunited with her birth family. This shows how genetic testing is being used in remarkable ways outside of the clinical laboratory. (Photo copyright: Highsmith family/People.)
Thanksgiving Reunion
Back in 1971, Melissa’s mother, Alta Apantenco, placed an advertisement in a local newspaper in Fort Worth, Texas, to hire a babysitter to care for her 21-month-old daughter. Apantenco hired Ruth Johnson to babysit her daughter without meeting the woman in person. Because Apantenco had to be at work, the child was handed over to Johnson by Apantenco’s roommate. The babysitter then allegedly abducted Melissa and disappeared with her.
Melissa’s family reported her missing to the police and searched for the snatched baby for more than 51 years. The family even organized a Facebook page called “Finding Melissa Highsmith” and sought outside assistance from the National Center for Missing and Exploited Children (NCMEC) in locating their lost relative, according to the New York Post.
The police and the FBI also got involved in the case, but few leads emerged over the decades.
Then, in September of 2022, Melissa’s family received a new lead regarding her location based on her father’s 23andMe DNA test results. Those results, along with a birthmark and date of birth, confirmed that Melissa was alive and well and residing in the Charleston, South Carolina area.
Over Thanksgiving weekend, Melissa was reunited with her mother, her father Jeffrie Highsmith, and two of her four siblings at a church in Fort Worth. She hopes to meet her remaining two siblings over the Christmas holidays.
“I can’t describe my feelings. I’m so happy to see my daughter that I didn’t ever think I would see again,” Apantenco told Saint Paul, Minnesota, television station KSTP.
“I couldn’t stop crying,” said Melissa’s sister Victoria Garner in a family statement. “I was overjoyed, and I’m still walking around in a fog trying to comprehend that my sister [was] right in front of me and that we found her,” The Guardian reported.
Clinical Laboratory Scientist Aids in the Investigation
The 23andMe test results alerted the family to the existence of a few unknown relatives that could be connected to the DNA of Melissa’s father. The family then contacted a genealogist and clinical laboratory scientist from Minnesota named Lisa Jo Schiele to help them interpret the results and potentially locate the missing woman. Schiele compared the DNA results with public records to help find Melissa Highsmith.
“I was able to use what we call traditional genealogy to find marriage records and things like that to find where Melissa was right now,” Schiele told KSTP. “At first glance, you look at these matches, but I’m like, ‘Holy cow, is this too good to be true?’ I’m very happy to help them navigate all of this.”
One of Melissa’s sisters, Sharon Highsmith, stated that her mother experienced deep feelings of guilt after Melissa’s abduction and had even faced accusations that she had something to do with the disappearance of her daughter.
“My mom did the best she could with the limited resources she had. She couldn’t risk getting fired, so she trusted the person who said they’d care for her child,” Sharon said in a family statement. “I’m grateful we have vindication for my mom,” The Guardian reported.
“I keep having to pinch myself to make sure I’m awake,” Melissa, who now resides in Fort Worth, told KSTP.
“It’s a miracle,” Apantenco said.
“A Christmas miracle,” Melissa added.
Due to the statute of limitations, which expired 20 years after Melissa turned 18, the babysitter who allegedly took Melissa cannot be criminally prosecuted.
“I’m angry our family was robbed for 51 years,’’ Melissa told Fort Worth news station WFAA.
This remarkable story illustrates how clinical laboratory skills combined with genetic testing results can be used outside of medical laboratory testing purposes to aid in solving criminal cases and other mysteries involving missing people.
Further advances in DNA testing combined with genetic databases that include DNA from greater numbers of people could result in more reunions involving missing persons who were identified because of genetic matching.
