From DNA to Disease: How Genetics, AI, and Personalized Medicine Are Changing Health Research

Scientists are learning that our bodies are more complex than we once thought. Two people can have the same genetic change but experience very different effects. A harmful germ might make one baby seriously ill while another stays healthy. Two women of the same age might also have very different risks of developing breast cancer.

New research is helping scientists understand why these differences happen.

This week, we look at studies on genetics, artificial intelligence (AI), new medicines, infant health, and cancer screening. Together, they show how scientists are finding better ways to understand diseases and develop treatments for different people.

The Same Gene Can Work Differently in Different Cells

Every cell in your body contains DNA, which carries the instructions for how your body grows and works. But not every cell uses those instructions in the same way.

For example, immune cells help protect us from infections. Different types of immune cells have different jobs. Some attack infected cells, while others help clean up harmful germs.

Two large studies published in Nature explored how genetic differences affect these cells.

In the first study, researchers examined 3.5 million immune cells from 1,042 people. They looked at which parts of DNA were open and available for cells to use.

They found more than 243,000 connections between genetic differences and how cells access DNA. About 60% of these connections were found only in certain types of immune cells.

The second study examined more than 5 million cells from 1,925 people. Researchers found that common genetic differences were linked to the activity of most genes they studied. About half of these effects appeared in only one or a few immune cell types.

Why is this important?

Imagine looking at the average test scores for an entire school. That average might hide the fact that some classrooms are doing very well while others need extra help.

The same thing happens when scientists study large groups of cells. An average result can hide important differences between individual cell types.

Researchers also discovered that a cell’s activity can change how its genes behave. For example, a genetic change linked to inflammatory bowel disease may have different effects depending on what an immune cell is doing.

These discoveries could help scientists better understand how certain genetic changes contribute to disease. However, finding a connection between a gene and a disease does not prove that the gene causes it.

Repeated Pieces of DNA May Help Explain Disease

Not all DNA looks like a simple set of instructions. Some parts contain short sequences that repeat over and over.

Scientists call these tandem repeats.

Think of a sentence in which the same word appears several times. One person might have the word repeated five times, while another has it repeated ten times.

Something similar happens in our DNA. Different people can have different numbers of these repeated sections.

For years, these parts of DNA were difficult to study. Now, better technology is helping researchers understand what they do.

In a new study, scientists examined more than 1.7 million regions of repeated DNA. They also studied over 5.4 million blood cells.

The researchers found more than 69,000 connections between repeated DNA and gene activity. About 31% of these connections were specific to certain immune cell types.

They also identified 1,490 repeated DNA regions that might directly affect how genes work.

These findings suggest that repeated DNA may help control important activities inside cells.

Scientists still need to learn whether these changes cause diseases. But the research provides new clues about how our DNA influences our health.

It also shows the power of studying several parts of a cell at once, including its DNA and the genes it is actively using.

Can AI Help Scientists Discover New Medicines?

Creating a new medicine takes years of work. Scientists often need to test thousands of chemicals before finding one that might treat a disease.

Artificial intelligence could speed up the early steps.

A study published in Science Advances tested a new system called FSGLD, which uses an AI model named DRUG-GAN.

The system helps scientists design possible drug molecules. These are tiny groups of atoms that could become ingredients in future medicines.

Researchers can use the system to create new molecules, design molecules aimed at a particular target in the body, or make groups of similar molecules.

The computer then predicts how these molecules might interact with their targets.

Think of a lock and key. A drug molecule needs to fit the right target in the body to produce the desired effect. Computer models can help scientists search for possible matches before testing them in a laboratory.

The researchers tested their system on three different types of drug targets and reported promising results.

They also developed a method that reduced the time needed for some computer calculations by 80% to 90% without reducing accuracy in their tests.

That could allow scientists to explore more possible drugs using less time and computing power.

But there is an important limitation.

A molecule designed by AI is not automatically a safe or effective medicine.

Scientists still need to test these molecules in laboratories, study their safety, and eventually conduct human trials before they can become approved treatments.

AI could help researchers find promising candidates faster, but it cannot replace careful scientific testing.

A New Clue About a Dangerous Disease in Premature Babies

Premature babies face special health risks because their bodies are still developing.

One serious condition is called necrotizing enterocolitis, or NEC. This disease damages the intestines and can become life-threatening.

Scientists have been trying to understand why NEC develops and how to prevent it.

A new study points to a possible cause in some cases: a toxin made by certain bacteria.

Researchers found a type of bacteria called Clostridium perfringens in premature infants who had severe or fatal cases of NEC.

Some of these bacteria carried a toxin called Delta toxin.

The researchers then studied what happened when mice were exposed to the bacteria or toxin.

The mice developed serious intestinal damage. Some also experienced brain injuries.

Scientists discovered that Delta toxin uses a protein called CD9 to enter certain cells. The toxin was especially harmful to macrophages, immune cells that fight infections and remove damaged material.

When these cells were affected, they triggered strong inflammation and a type of cell death. This also contributed to damage in the protective barrier that normally helps keep harmful substances out of the brain.

Researchers found that mice without CD9 in certain cells were protected from several harmful effects of the toxin. An antibody designed to block the toxin also protected mice.

These results are encouraging, but they do not mean doctors have found a new treatment for NEC.

Much of the research was done in mice, not human babies.

Scientists will need to conduct more studies to understand whether blocking this toxin could safely prevent or treat NEC in premature infants.

Still, the discovery gives researchers a possible new target for future treatments.

Could Breast Cancer Screening Be Based on Personal Risk?

Breast cancer screening helps doctors find cancer early, sometimes before symptoms appear.

But not everyone has the same risk of developing breast cancer.

Age is one risk factor, but family history, breast density, and genetics can also affect a person’s chances.

Researchers in Singapore studied whether combining these factors could help doctors make better screening recommendations.

The study, called BREATHE, included 4,513 women.

Scientists used a tool called a polygenic risk score. This score looks at many small genetic differences to estimate a person’s risk of developing a disease.

They combined this score with other information, including family history, breast density, and a commonly used breast cancer risk model.

The researchers then explored how this information might change screening recommendations.

For women ages 40 to 49, many would have been placed into groups needing mammograms every two years instead of every year. Some would have been assigned to a group where screening was not recommended during that period.

The combined approach also identified more cancer cases than the genetic score alone.

This suggests that using several types of health information could help doctors develop more personalized screening plans.

However, the study followed relatively few women who developed breast cancer during the research period.

Scientists need larger studies to test whether these recommendations are safe and effective.

For now, the findings should not be used as a reason to skip recommended cancer screenings.

The Future of Medicine Is Becoming More Personal

These studies share an important idea: people and cells do not always respond in the same way.

A genetic change might affect one type of immune cell but not another. Repeated sections of DNA may change how genes work. A bacterial toxin might help explain some cases of a dangerous infant disease.

Meanwhile, AI could help scientists discover new drugs, and genetic information could eventually help doctors choose better screening plans.

The common goal is to move away from treating everyone as if their biology were exactly the same.

Scientists call this precision medicine. It means using information about a person’s genes, health, and other characteristics to make more informed decisions about prevention and treatment.

But there is still a long road between a scientific discovery and a treatment that helps patients.

Researchers must confirm their findings, test new approaches carefully, and make sure they work safely in real-world settings.

The more scientists learn about the differences between cells and people, the better their chances of developing treatments and prevention strategies that fit individual needs.

Researchers also discovered that a cell’s activity can change how its genes behave. For example, a genetic change linked to inflammatory bowel disease may have different effects depending on what an immune cell is doing.

These discoveries could help scientists better understand how certain genetic changes contribute to disease. However, finding a connection between a gene and a disease does not

Wrapping Up

It’s a medicine-heavy week, but that’s one thing we all share: a design for health and wellness.