Advances in AI and CRISPR Technologies To Expedite Our Understanding of Genetic Diseases

There are as many as 7,000 rare genetic diseases that occur due to genetic disorders. These disorders happen when a mutation occurs in our genes, though it doesn’t always end up in a disease.

Genes are made of deoxyribonucleic acid or DNA, which contains instructions regarding our body’s characteristics and functions, and a change in the DNA sequence causes a genetic disorder.

Because we receive half of our genes from each biological parent, we may inherit a gene mutation from either or both. Such disorders could also be a result of a combination of gene mutations, damage to chromosomes, which are thread-like structures that carry genes, and environmental factors.

This intricate genetic landscape, marked by the complexity, low prevalence, and hidden nature of disorders, significantly complicates the process of achieving an accurate diagnosis. In many cases, rare genetic diseases could be due to a broad range of variants in genes that could be unique to an individual, making it even more difficult to identify the specific cause of a patient’s symptoms.

Moreover, the diagnosis process also requires extensive genetic analyses and time to compare the genetic makeup of the patient with known disease patterns. So, not only is the diagnosis challenging, but it may take a long time too, which, in turn, delays the treatment. A new Danish study, however, is all set to make the diagnosis of such diseases more efficient and faster.

Published in The American Journal of Human Genetics, the study made use of the CRISPR tech to switch on genes in cells that are easily accessible, such as blood or skin. Then, the researchers measured whether messenger RNA is assembled correctly in a molecular biology process called splicing.

This innovation addresses a significant hurdle in genetic research. This is a big development because 19% of genes that are with the disease are not active in these readily accessible tissues like blood cells and skin cells. This makes it even harder to get a clear picture of the genetic basis of diseases.

According to experts, this development could revolutionize genetic diagnostics. 

“This means that until now, we have not been able to properly investigate the genes because they are only active in specific tissues, such as the nervous system, and this is a major obstacle when it comes to understanding whether a particular gene variant is the cause of a patient’s disease.” 

– Uffe Birk Jensen, a Clinical Professor and Chair at Aarhus University’s Department of Clinical Medicine – Department for Clinical Genetics

The activation of CRISPR here presents a big opportunity. CRISPR, which stands for “Clustered Regularly Interspaced Short Palindromic Repeats,” is a genome editing technology that is used to modify DNA sequences selectively and precisely. It consists of the Cas9 enzyme, which is used as molecular scissors to cut the DNA. Then there’s guide-RNA, which recognizes the target that needs to be modified and then directs Cas9 to that region of the DNA.

Using this technique, researchers “turned on” genes that are normally not active in accessible cells, including the MPZ gene—typically active only in nerve pathways—now activated in skin cells. This breakthrough offers new possibilities for understanding and diagnosing genetic diseases.

According to Uffe Birk Jensen, CRISPR activation has allowed the gene to be turned on in a natural environment, which means the gene modification doesn’t have to be in cell models but rather can use a patient’s sample. He added:

“The same method can be used for different patients and easily adapted to other genes, and the advantage is that it’s very fast with the possibility of results within a few weeks.” 

Researchers are hoping to use this method for more accessible, effective, and accurate diagnosis of genetic diseases. It has already been introduced in the clinic so that “the technology can contribute to making the right diagnosis when we find possible splicing variants,” said the study co-author Thorkild Terkelsen, who’s a postdoc at the Department of Biomedicine.

But this is not all. The team is further exploring the broader potential of the technique and making adjustments so it becomes “even easier to use in the clinic,” Terkelsen added.

The study, which is a collaborative effort between the Department of Clinical Genetics and the Department of Molecular Medicine of Aarhus University Hospital, received funding from the Independent Research Fund Denmark. 

The Vast Potential of Latest Technologies in Understanding Genetic Diseases

While CRISPR came into existence in 1987, it has only recently started becoming popular. This promising, innovative technology offers researchers and scientists an effective way to edit DNA structures and change gene function. 

The technique has several possible uses, including treating and reducing the spread of disorders as well as editing inherited deficiencies. When it comes to gene therapy, CRISPR/Cas9 technology offers some great potential by genetically editing a patient-specific mutation that is otherwise not treatable with traditional treatments. So, by alleviating complicated diseases, CRISPR technology aims to provide healthier lives for people. 

As CRISPR is now increasingly being used for human disorder rectification, attempts are being made to reduce off-targets and allow for more precision. Here, artificial intelligence, in particular, is helping genome editing achieve not just more precision but also efficiency and affordability in tackling various diseases. 

AI is the simulation of human intelligence in robots to think, learn, and predict like humans. The technology aims to transform many industries, including healthcare. 

Recently, we talked about how AI’s ability to recognize trends within a set of variables is advancing healthcare exponentially. By utilizing AI algorithms, researchers are able to successfully predict sepsis infection in high-risk patients before the onset of symptoms and save lives. In another case, AI-powered prediction models analyze genomic data to predict disease with high accuracy.

This wide-ranging branch of computer science is proving to be a powerful tool in the detection of heredity and gene-related disorders. In gene research, AI’s highly versatile branch, deep learning, is being increasingly leveraged. 

Based on artificial neural networks, deep learning teaches computers to process large amounts of data. As a subset of machine learning, a deep learning model attempts to simulate the behavior of the human brain by learning from the provided data. 

Deep learning involves multiple layers of interconnected nodes to create deep neural networks that can learn complex representations of data by discovering hierarchical patterns in it. Based on the patterns, it then produces accurate insights and predictions. Such algorithms can learn and improve from data on their own without requiring manual interference. 

Deep learning models have been achieving great performance in biomedical applications, with studies suggesting that such models can perform just as well and maybe even better than humans at key healthcare tasks like diagnosing diseases. After all, AI can reduce human errors, process vast amounts of data, and augment knowledge capacity.

Given the complexity of data in healthcare, AI can be efficiently applied for diagnosis, treatment recommendations, administrative tasks, and patient engagement across various healthcare domains. 

However, as we noted in our ‘AI-Powered Prediction Models Set to Advance HealthCare by Leaps and Bounds‘ piece, mistakes can occur when AI is applied in the real world, and we need to be mindful of that.

The Promise of AI in Researching, Diagnosing, & Treating Genetic Disorders

Utilizing AI for understanding, diagnosing, and treating disease isn’t anything new. Rule-based systems have demonstrated promise for accurately diagnosing and treating disease. But in recent years, advancement in technology has allowed it to be done in regards to genetic diseases and with improved results. 

For instance, IBM’s Watson has been gaining a lot of traction for its focus on precision medicine, particularly cancer diagnosis and treatment. For this, Watson combines machine learning and NLP capabilities. 

As for genetic therapy, AI tools like DeepCRISPR, CRISTA, and DeepHF are being used to guide RNAs for a specified target sequence. Predicting optimal guide RNAs takes multiple factors into account, such as genomic context, desired mutation type, on-target scores, off-target scores, prospective off-target locations, and the possible impacts of gene editing on its function.

AI models further help optimize different technologies to edit the genome, such as base (making targeted changes to the sequence of a piece of DNA), prime (a ‘search-and-replace’ editing), and epigenome (adding or removing molecular markers from DNA), which allows for precise and programmable changes to DNA sequences without relying on the donor DNA templates. 

Moreover, AI’s collaboration with genome editing allows for personalized treatments tailored to genetic profiles, analyzing patients’ genomic data to identify mutations and disease-associated biomarkers, such as those for Alzheimer’s and cancer, early on, leading to customized care.

Scientists and researchers believe by helping clinicians determine the root cause of diseases quicker, technology like AI allows patients to receive the right treatment sooner, significantly improving patient care.

Besides allowing for faster diagnosis, AI is also doing it with more accuracy. Back in Oct. 2021, a study conducted by University of Utah Health researchers and led by Rady Children’s Hospital in San Diego developed a Fabric GEM algorithm that leverages AI to find errors in DNA that lead to rare genetic disorders in babies. 

On being tested, GEM was able to identify the causative gene as one of its top two candidates 92% of the time, outperforming existing tools that were successful only 60% of the time. Moreover, unlike other tools, GEM could also find “structural variants,” which are larger and more complex, as causes of disease. 

For this, GEM learned from vast and ever-growing knowledge and then cross-referenced it with massive datasets of genomic sequences from different groups of the population and all the information about the clinical disease, which was then combined with the patient’s genome sequence and medical records. For assistance, GEM can be coupled with an NLP tool to scan loads of clinical notes.

Most recently, Google’s DeepMind introduced a new AI model that can predict genetic diseases. Named AlphaMissense, the language model is trained on protein sequences and predicts the structures of hundreds of millions of proteins from their amino acid composition.

A “pathogenicity score” between 0 and 1 has been assigned by the model to each of the 71 million possible missense variants. This is based on its knowledge of the effects of other closely related mutations. Here, the higher the score, the more likely a particular mutation is to cause or be associated with disease. 

DeepMind researchers claim 90% accuracy for AlphaMissense, with 89% of variants classified. With AlphaMissense, the idea is to help accelerate research into genetic variants, improve disease diagnosis, and find new treatments faster. It can also help doctors quickly rule out any other potential genetic mutations in a patient’s DNA and be sure that they’re giving the right treatment.

Companies Leveraging AI & CRISPR Technologies

Now, let’s look at a couple of prominent companies in the healthcare sector that are making use of AI and CRISPR to innovate and develop solutions for the betterment of human lives and society. 

#1. CRISPR Therapeutics

The Zug-headquartered company, a pioneer in employing CRISPR technology for treating both common and rare diseases, has secured IP rights for CRISPR/Cas9 and related technologies. Following this strategic move, Crispr Therapeutics received regulatory approval in Q4 of 2023 for CASGEVY to treat sickle cell disease (SCD) in the US and for SCD and transfusion-dependent beta-thalassemia (TDT) in the UK and Bahrain.

Leveraging these approvals, CEO Samarth Kulkarni stated in 2024:

“We will continue to drive forward our programs and expand our pipeline with the goal of delivering paradigm-shifting gene editing therapies to patients. We are well positioned to execute our clinical trials across various therapeutic areas, including oncology, autoimmune, cardiovascular, and diabetes.”

Reflecting the company’s advancements and market confidence, Crispr Therapeutics, with a market cap of $5 bln, sees its shares trading at $62.95, marking a 0.56% increase year-to-date (YTD). The company reported a revenue of $170 mln over the trailing twelve months (TTM), with an EPS (TTM) of -4.48 and a P/E (TTM) ratio of -14.05. Demonstrating investor confidence, in Jan. 2024, Cathie Wood’s ARK Investment acquired 167K shares of Crispr Therapeutics.

Click here to learn all about investing in CRISPR Therapeutics.

#2. UnitedHealth Group Incorporated

This diversified healthcare company has been making a lot of use of AI, machine learning (ML), and natural language processing (NLP), with its COO Dirk McMahon noting:

“Long-term, we think there’s great hope for those.”

Over a decade ago, UnitedHealth Group acquired Optum, which uses technology and AI for decision-making and predictive analysis. In its Q4 earnings, the company reported a 24.2% increase in the Optum division’s revenue to $59.5 bln. A few years ago, the company launched a Level 2 digital therapy platform for diabetes patients and also incorporated a virtual assistance platform to optimize customer support.

With a market cap of $473.32 bln, the shares of UnitedHealth Group are currently trading at $511.74, down 2.8% YTD. The company posted a revenue (TTM) of $317.62 bln and has an EPS (TTM) of 23.85 and a P/E (TTM) of 21.46.

Concluding Thoughts

CRISPR has emerged as a simple and rapid genome editing technology that has great potential in treating genetic disorders. However, several challenges like high costs and off-target editing persist, which calls for the need to have improved editing efficiency while ensuring safety.

Besides CRISPR, AI is another technology changing the face of healthcare with its unparalleled ability to recognize trends. Together, AI and CRISPR technologies are unlocking exciting possibilities for quick understanding of genes and improving medical treatments for genetic diseases.

Click here to learn about the top CRISPR companies to invest in.