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.
Cellular healthcare is an approach that goes beyond clinical laboratory testing to identify the location of specific cancer cells and aid in treatment decisions
Advances in synthetic biology and genetic engineering are leading to development of bacterial biosensors that could eventually aid pathologists and clinical laboratories in diagnosis of many types of cancers.
One recent example comes from researchers at the University of California San Diego (UCSD) who worked with colleagues in Australia to engineer bacteria that work as “capture agents” and bind to tumorous material.
The KRAS gene is associated with colorectal cancer. The researchers named their development the Cellular Assay for Targeted CRISPR-discriminated Horizontal gene transfer (CATCH).
CATCH successfully detected cancer in the colons of mice. The researchers believe it could be used to diagnose cancers, as well as infections and other diseases, in humans as well, according to a UCSD news release.
“If bacteria can take up DNA, and cancer is defined genetically by a change in its DNA, then, theoretically, bacteria could be engineered to detect cancer,” gastroenterologist Daniel Worthley, PhD, a cancer researcher at Colonoscopy Clinic in Brisbane, Australia, told MedicalResearch.com. This research could eventually provide clinical laboratories and anatomic pathologists with new tools to use in diagnosing certain types of cancer. (Photo copyright: Colonoscopy Clinic.)
Tapping Bacteria’s Natural Competence
In their Science paper, the researchers acknowledged other synthetic biology achievements in cellular biosensors aimed at human disease. But they noted that more can be done by leveraging the “natural competence” skill of bacteria.
“Biosensors have not yet been engineered to detect specific extracellular DNA sequences and mutations. Here, we engineered naturally competent Acinetobacter baylyi (A. baylyi) to detect donor DNA from the genomes of colorectal cancer cells, organoids, and tumors,” they wrote.
“Many bacteria can take up DNA from their environment, a skill known as natural competence,” said Rob Cooper, PhD, co-first author of the study and a scientist at US San Diego’s Synthetic Biology Institute, in the news release. A. baylyi is a type of bacteria renowned for success in doing just that, the NCI article pointed out.
This enabled them to explore “free-floating DNA sequences on a genomic level.”
Those sequences were compared to “known cancer DNA sequences.”
A. baylyi (genetically modified) was tested on its ability to detect “mutated and healthy KRAS DNA.”
Only bacteria that had “taken up mutated copies of KRAS … would survive treatment with a specific drug.”
“It was incredible when I saw the bacteria that had taken up the tumor DNA under the microscope. The mice with tumors grew green bacterial colonies that had acquired the ability to be grown on antibiotic plates,” said Josephine Wright, PhD, Senior Research Fellow, Gut Cancer Group, South Australian Health and Medical Research Institute (SAHMRI), in the news release.
Detecting DNA from Cancer Cells In Vitro and in Mice
Findings in vitro and in mice include the following:
The engineered bacteria enabled detection of DNA with KRAS G12D from colorectal cancer cells made in the lab, NCI reported.
When mice were injected with colorectal cancer cells, the researchers’ technology found tumor DNA, Engadget reported.
The study adds to existing knowledge of horizontal gene transfer from bacteria to bacteria, according to UCSD.
“We observed horizontal gene transfer from the tumor to the sensor bacteria in our mouse model of colorectal cancer. This cellular assay for targeted, CRISPR-discriminated horizontal gene transfer (CATCH) enables the biodetection of specific cell-free DNA,” the authors wrote in Science.
“Colorectal cancer seemed a logical proof of concept as the colorectal lumen is full of microbes and, in the setting of cancer, full of tumor DNA,” gastroenterologist Daniel Worthley, PhD, a cancer researcher at Colonoscopy Clinic in Brisbane, Australia, told MedicalResearch.com.
Finding More Cancers and Treatment
More research is needed before CATCH is used in clinical settings. The scientists are reportedly planning on adapting CATCH to multiple bacteria that can locate other cancers and infections.
“The most exciting aspect of cellular healthcare … is not in the mere detection of disease. A laboratory can do that,” wrote Worthley in The Conversation. “But what a laboratory cannot do is pair the detection of disease (a diagnosis) with the cells actually responding to the disease [and] with appropriate treatment.
“This means biosensors can be programmed so that a disease signal—in this case, a specific sequence of cell-free DNA—could trigger a specific biological therapy, directly at the spot where the disease is detected in real time,” he added.
Clinical laboratory scientists, pathologists, and microbiologists may want to stay abreast of how the team adapts CATCH, and how bacterial biosensors in general continue to develop to aid diagnosis of diseases and improve ways to target treatment.
Expanded genomic dataset includes a wider racial diversity which may lead to improved diagnostics and clinical laboratory tests
Human genomic research has taken another important step forward. The National Institutes of Health’s All of Us research program has reached a milestone of 250,000 collected whole genome sequences. This accomplishment could escalate research and development of new diagnostics and therapeutic biomarkers for clinical laboratory tests and prescription drugs.
The NIH’s All of Us program “has significantly expanded its data to now include nearly a quarter million whole genome sequences for broad research use. About 45% of the data was donated by people who self-identify with a racial or ethnic group that has been historically underrepresented in medical research,” the news release noted.
Detailed information on this and future data releases is available at the NIH’s All of us Data Roadmap.
“For years, the lack of diversity in genomic datasets has limited our understanding of human health,” said Andrea Ramirez, MD, Chief Data Officer, All of Us Research Program, in the news release. Clinical laboratories performing genetic testing may look forward to new biomarkers and diagnostics due to the NIH’s newly expanded gene sequencing data set. (Photo copyright: Vanderbilt University.)
Diverse Genomic Data is NIH’s Goal
NIH launched the All of Us genomic sequencing program in 2018. Its aim is to involve more than one million people from across the country and reflect national diversity in its database.
So far, the program has grown to include 413,450 individuals, with 45% of participants self-identifying “with a racial or ethnic group that has been historically under-represented in medical research,” NIH said.
“By engaging participants from diverse backgrounds and sharing a more complete picture of their lives—through genomic, lifestyle, clinical, and social environmental data—All of Us enables researchers to begin to better pinpoint the drivers of disease,” said Andrea Ramirez, MD, Chief Data Officer of the All of Us research program, in the news release.
More than 5,000 researchers are currently registered to use NIH’s All of Us genomic database. The vast resource contains the following data:
245,350 whole genome sequences, which includes “variation at more than one billion locations, about one-third of the entire human genome.”
1,000 long-read genome sequences to enable “a more complete understanding of the human genome.”
Analysis of drugs effectiveness in different patients.
Data Shared with Participants
Participants in the All of Us program, are also receiving personalized health data based on their genetic sequences, which Dark Daily previously covered.
“Through a partnership with participants, researchers, and diverse communities across the country, we are seeing incredible progress towards powering scientific discoveries that can lead to a healthier future for all of us,” said Josh Denny, MD, Chief Executive Officer, All of Us Research Program, in the news release.
“[Researchers] can get access to the tools and the data they need to conduct a project with our resources in as little as two hours once their institutional data use agreement is signed,” said Fornessa Randal, Executive Director, Center for Asian Health Equity, University of Chicago, in a YouTube video about Researcher Workbench.
For diagnostics professionals, the growth of available whole human genome sequences as well as access to participants in the All of Us program is noteworthy.
Also impressive is the better representation of diversity. Such information could result in medical laboratories having an expanded role in precision medicine.
Findings could lead to deeper understanding of why we age, and to medical laboratory tests and treatments to slow or even reverse aging
Can humans control aging by keeping their genes long and balanced? Researchers at Northwestern University in Evanston, Illinois, believe it may be possible. They have unveiled a “previously unknown mechanism” behind aging that could lead to medical interventions to slow or even reverse aging, according to a Northwestern news release.
Should additional studies validate these early findings, this line of testing may become a new service clinical laboratories could offer to referring physicians and patients. It would expand the test menu with assays that deliver value in diagnosing the aging state of a patient, and which identify the parts of the transcriptome that are undergoing the most alterations that reduce lifespan.
It may also provide insights into how treatments and therapies could be implemented by physicians to address aging.
“I find it very elegant that a single, relatively concise principle seems to account for nearly all of the changes in activity of genes that happen in animals as they change,” Thomas Stoeger, PhD, postdoctoral scholar in the Amaral Lab who led the study, told GEN. Clinical laboratories involved in omics research may soon have new anti-aging diagnostic tests to perform. (Photo copyright: Amaral Lab.)
Possible ‘New Instrument’ for Biological Testing
Researchers found clues to aging in the length of genes. A gene transcript length reveals “molecular-level changes” during aging: longer genes relate to longer lifespans and shorter genes suggest shorter lives, GEN summarized.
The phenomenon the researchers uncovered—which they dubbed transcriptome imbalance—was “near universal” in the tissues they analyzed (blood, muscle, bone, and organs) from both humans and animals, Northwestern said.
According to the National Human Genome Research Institute fact sheet, a transcriptome is “a collection of all the gene readouts (aka, transcript) present in a cell” shedding light on gene activity or expression.
The Northwestern study suggests “systems-level” changes are responsible for aging—a different view than traditional biology’s approach to analyzing the effects of single genes.
“We have been primarily focusing on a small number of genes, thinking that a few genes would explain disease,” said Luis Amaral, PhD, Senior Author of the Study and Professor of Chemical and Biological Engineering at Northwestern, in the news release.
“So, maybe we were not focused on the right thing before. Now that we have this new understanding, it’s like having a new instrument. It’s like Galileo with a telescope, looking at space. Looking at gene activity through this new lens will enable us to see biological phenomena differently,” Amaral added.
In their Nature Aging paper, Amaral and his colleagues wrote, “We hypothesize that aging is associated with a phenomenon that affects the transcriptome in a subtle but global manner that goes unnoticed when focusing on the changes in expression of individual genes.
“We show that transcript length alone explains most transcriptional changes observed with aging in mice and humans,” they continued.
In tissues studied, older animals’ long transcripts were not as “abundant” as short transcripts, creating “imbalance.”
“Imbalance” likely prohibited the researchers’ discovery of a “specific set of genes” changing.
As animals aged, shorter genes “appeared to become more active” than longer genes.
In humans, the top 5% of genes with the shortest transcripts “included many linked to shorter life spans such as those involved in maintaining the length of telomeres.”
Conversely, the researchers’ review of the leading 5% of genes in humans with the longest transcripts found an association with long lives.
Antiaging drugs—rapamycin (aka, sirolimus) and resveratrol—were linked to an increase in long-gene transcripts.
“The changes in the activity of genes are very, very small, and these small changes involve thousands of genes. We found this change was consistent across different tissues and in different animals. We found it almost everywhere,” Thomas Stoeger, PhD, postdoctoral scholar in the Amaral Lab who led the study, told GEN.
In their paper, the Northwestern scientists noted implications for creation of healthcare interventions.
“We believe that understanding the direction of causality between other age-dependent cellular and transcriptomic changes and length-associated transcriptome imbalance could open novel research directions for antiaging interventions,” they wrote.
While more research is needed to validate its findings, the Northwestern study is compelling as it addresses a new area of transcriptome knowledge. This is another example of researchers cracking open human and animal genomes and gaining new insights into the processes supporting life.
For clinical laboratories and pathologists, diagnostic testing to reverse aging and guide the effectiveness of therapies may one day be possible—kind of like science’s take on the mythical Fountain of Youth.
Viruses are between 27,000 to 48,500 years old and not dangerous, but researchers say thawing permafrost may one day release pathogens capable of infecting humans
Last fall, European researchers working with virologists and genetic scientists at the Aix-Marseille University in France reported having revived and characterized 13 previously unknown “zombie” viruses isolated from Siberian permafrost samples, including one that was almost 50,000 years old. This will be of particular interest to microbiologists and clinical laboratory managers since these organisms are new to science and may be precursors to infectious agents active in the world today.
The work of the European scientists demonstrates how advancements in genome sequencing and analysis of DNA data are becoming, faster, less expensive, and more precise. That’s good because the researchers warned that, should the permafrost continue to thaw, other previously dormant viruses could be released, posing potential risks for public health.
The pathogens isolated by the researchers are so-called “giant viruses” that infect Acanthamoeba, a commonly found genus of amoeba, and thus are not likely to pose an immediate health threat, the researchers wrote.
However, the scientists expressed concern. “We believe our results with Acanthamoeba-infecting viruses can be extrapolated to many other DNA viruses capable of infecting humans or animals. It is thus likely that ancient permafrost … will release these unknown viruses upon thawing,” they stated in their Viruses paper.
It’s unknown how long the viruses “could be infectious once exposed to outdoor conditions (UV light, oxygen, heat), and how likely they will be to encounter and infect a suitable host in the interval,” they added. However, “the risk is bound to increase in the context of global warming, in which permafrost thawing will keep accelerating, and more people will populate the Arctic in the wake of industrial ventures.”
“In nature we have a big natural freezer, which is the Siberian permafrost,” virologist Paulo Verardi, PhD (above), head of the Department of Pathobiology and Veterinary Science at the University of Connecticut, told The Washington Post. “And that can be a little bit concerning.” However, “if you do the risk assessment, this is very low. We have many more things to worry about right now.” Nevertheless, clinical laboratories may want to remain vigilant. (Photo copyright: University of Connecticut.)
Extremely Old, Very Large Viruses
The newly discovered viruses were found in seven different permafrost samples. Radiocarbon dating determined that they had been dormant for 27,000 to 48,500 years. But viruses contained in permafrost could be even older, the researchers wrote, as the time limit is “solely dictated by the validity range of radiocarbon dating.”
In their Viruses paper, the researchers noted that most of the 13 viruses are “at a preliminary stage of characterization,” and others have been isolated in the research laboratory “but not yet published, pending their complete genome assembly, annotation, or detailed analysis.”
“Every time we look, we will find a virus,” study co-author Jean-Michel Claverie, PhD, told The Washington Post. “It’s a done deal. We know that every time we’re going to look for viruses—infectious viruses in permafrost—we are going to find some.”
Claverie is a professor emeritus of genomics and bioinformatics in the School of Medicine at Aix-Marseille Université in Marseille, France. He leads a university laboratory known for its work in “paleovirology,” and in 2003, discovered the first known giant virus, dubbed Mimivirus. The research team included scientists from Germany and Russia.
According to CNN, unlike regular viruses that generally require an electron microscope to be viewed, giant viruses can be seen under a standard light (optical) microscope. Claverie’s laboratory previously isolated giant viruses from permafrost in 2014 and 2015.
Protecting Against Accidental Infection
To demonstrate the infectious potential of the viruses, the researchers inserted the microbes into cultured amoeba cells, which the researchers describes as “virus bait,” The Washington Post reported. One advantage of using Acanthamoeba cultures is to maintain “biological security,” the researchers wrote in their paper.
“We are using [the amoeba’s] billion years of evolutionary distance with human and other mammals as the best possible protection against an accidental infection of laboratory workers or the spread of a dreadful virus once infecting Pleistocene mammals to their contemporary relatives,” the paper noted. “The biohazard associated with reviving prehistorical amoeba-infecting viruses is thus totally negligible compared to the search for ‘paleoviruses’ directly from permafrost-preserved remains of mammoths, woolly rhinoceros, or prehistoric horses.”
The paper cites earlier research noting the presence of bacteria in ancient permafrost samples, “a significant proportion of which are thought to be alive.” These include relatives of contemporary pathogens such as:
“We can reasonably hope that an epidemic caused by a revived prehistoric pathogenic bacterium could be quickly controlled by the modern antibiotics at our disposal,” the researchers wrote, but “the situation would be much more disastrous in the case of plant, animal, or human diseases caused by the revival of an ancient unknown virus.”
However, according to The Washington Post, “Virologists who were not involved in the research said the specter of future pandemics being unleashed from the Siberian steppe ranks low on the list of current public health threats. Most new—or ancient—viruses are not dangerous, and the ones that survive the deep freeze for thousands of years tend not to be in the category of coronaviruses and other highly infectious viruses that lead to pandemics.”
Cornell University virologist Colin Parrish, PhD, President of the American Society for Virology, told The Washington Post that an ancient virus “seems like a low risk compared to the large numbers of viruses that are circulating among vertebrates around the world, and that have proven to be real threats in the past, and where similar events could happen in the future, as we still lack a framework for recognizing those ahead of time.”
Anthony Fauci, MD, former Director of the National Institute of Allergy and Infectious Diseases (NIAID), responded to an earlier study from Claverie’s lab by outlining all the unlikely events that would have to transpire for one of these viruses to cause a pandemic. “The permafrost virus must be able to infect humans, it must then [cause disease], and it must be able to spread efficiently from human to human,” he told The Washington Post in 2015. “This can happen, but it is very unlikely.”
Thus, clinical laboratories probably won’t see new diagnostic testing to identify ancient viruses anytime soon. But it’s always best to remain vigilant.
