Alsesai, 4Floor, SRI, No. 3639, OuYa Avenue, CBE, Xi'an (2026)

04/10/2026

The brain has always been one of the hardest places to develop new medicines.

It isn't just a collection of cells. It is a vast network of neurons, synapses and molecular signals — protected by the blood-brain barrier and constantly changing.

Now AI is beginning to change how scientists search for treatments.

A new generation of AI-driven drug discovery systems can analyze enormous amounts of biological data, identify promising molecular targets, and help design compounds that would be extremely difficult to discover through traditional trial-and-error approaches.

One recent example is GATC-1021, an AI-derived drug candidate developed for opioid use disorder. In preclinical studies, the compound reduced opioid intake in rats and was associated with changes in synaptic organization and gene expression in brain regions involved in reward.

Other AI-driven approaches are now being explored for neurological diseases ranging from Parkinson's disease to stroke.

But the most interesting part isn't simply that AI can design molecules.

It is that AI may eventually help us understand the brain as a system — identifying which cells, pathways and molecular switches should be changed, and then designing medicines specifically around those insights.

For decades, drug discovery often looked like:

Target → molecule → test → fail → repeat.

AI is pushing toward:

Biological data → prediction → target → molecule → experiment → feedback → better prediction.

That could create something much more powerful than an AI drug designer.

A learning system for discovering medicines.

We're still far from proving that AI can reliably create effective human medicines. Most candidates remain experimental, and no fully AI-designed drug has yet become an approved medicine.

But the direction is clear.

The next generation of medicines may not be discovered by searching through molecules.

They may be discovered by searching through biology.

.alsesai.com

24/09/2026

🧬 **Scientists Just Built a Computer Made Entirely of DNA!**

Researchers have taken a massive leap in bio-computing by creating a computer that uses DNA molecules instead of silicon chips to process data and perform complex calculations.

**Why this is a game-changer:**

* 🧠 **Unbelievable Density:** A single gram of DNA can theoretically store hundreds of petabytes of data.
* ⚡ **Parallel Processing:** DNA computers can perform billions of operations simultaneously, tackling complex algorithms in seconds.
* 🔬 **Biomedical Revolution:** Imagine tiny biological "smart processors" operating inside the human body to target diseases, deliver precise drug doses, or monitor health at a cellular level.

We are officially stepping into an era where biology and computer science are seamlessly merging. The future of tech isn't just digital—it's organic. 🌐✨

.com .alsesai.com

09/09/2026

# AI Just Mapped 9 Billion Possible Mutations in the Human Genome.

For decades, genetics has had a frustrating problem:

**We can find mutations much faster than we can understand them.**

A DNA sequence may differ from the reference genome by a single letter. But what does that one-letter change actually do?

Does it alter gene expression?

Does it disrupt RNA splicing?

Does it affect a regulatory element?

Or does it do nothing at all?

Now AI is beginning to tackle this problem at an entirely different scale.

Google DeepMind has released **AlphaGenome Atlas**, a massive predictive resource covering roughly **9 billion possible single-nucleotide changes across the human genome**.

Instead of simply cataloguing mutations, AlphaGenome attempts to predict their **biological consequences** — including effects on gene regulation, RNA processing, chromatin accessibility, and other molecular processes.

Think about the shift:

# # # Before:

**DNA sequencing → Find a mutation → Study it**

# # # Now:

**DNA → AI → Predict billions of possible mutations → Prioritize the important ones → Test them experimentally**

That changes the economics of biological discovery.

Scientists cannot realistically test billions of mutations one by one.

But an AI model can potentially **screen the search space first** and tell researchers where to look.

And this may be especially important for **rare genetic diseases**, where a patient's mutation may sit in a poorly understood region of the genome.

The bigger story isn't simply that AI analyzed a huge amount of DNA.

It's that biology may be moving from:

**“What mutations exist?”**

to

# # # **“What would happen if every possible mutation occurred?”**

That is a fundamentally different question.

AlphaFold helped make protein structures computationally searchable.

AlphaGenome is pushing toward something potentially broader:

**making the functional consequences of DNA variation searchable.**

Of course, predictions are not experiments. A model can generate hypotheses, but biology still has the final vote.

Yet if AI can predict the effects of billions of genetic changes before scientists test them…

# # # **Are we approaching a future where the human genome becomes searchable—not just as a sequence, but as a map of possible biological outcomes?**

.alsesai.com

08/09/2026

For decades, organ transplantation has faced the same brutal problem:

There are far more patients than human organs.

Now, scientists may have found a very unconventional way to buy patients more time.

A genetically modified pig kidney was transplanted into a man with end-stage kidney disease.

It worked for 271 days.

During that time, he did not need dialysis.

And then something even more remarkable happened:

He received a human kidney.

The pig kidney had served as a biological bridge while he waited for a human donor.

🧬 This wasn't an ordinary pig kidney.

The organ had been extensively genetically engineered.

Scientists removed pig genes associated with major immune incompatibilities, inserted seven human genes to improve compatibility, and inactivated porcine endogenous retroviruses to reduce infection risks.

In other words, scientists weren't simply putting a pig organ into a human.

They were trying to engineer a pig organ for the human body.

And that's where this story becomes much bigger than one patient.

Think about the current transplant system:

Kidney failure → dialysis → wait for a human donor

Now imagine:

Kidney failure → gene-edited pig kidney → no dialysis → wait for a human donor → human kidney

The pig kidney doesn't necessarily have to be the final destination.

It could become a temporary biological bridge.

In this first reported case, the patient later received a human kidney that functioned immediately, with no evidence that the pig transplant had sensitized him against the human donor. No porcine pathogen transmission was detected during follow-up.

And this isn't happening in isolation.

Five patients have received the same company's gene-edited pig kidney under expanded access since 2024, with three maintaining kidney function for more than eight months without dialysis.

But let's be clear:

This is not yet a replacement for human kidneys.

The current evidence is extremely limited, and scientists still need to solve long-term rejection, durability, infection risk, immunosuppression, and scalability.

Still, something important may have changed.

We are no longer asking only:

“Can an animal organ replace a human organ?”

We are beginning to ask:

“Can an engineered animal organ keep a human alive long enough to receive the organ they actually need?”

That is a very different question.

And perhaps the future of transplantation won't begin with a human donor.

It may begin with a pig. 🐷

.alsesai.com

05/09/2026

A single genetic typo can stop a cell from making a critical protein.

Scientists are now trying to **teach the cell to read past that mistake.**

Researchers at the University of Toronto, University Health Network and SickKids have developed an engineered form of **transfer RNA (tRNA)** that can bypass a premature “STOP” signal in genetic instructions.

The study was published in *Science* on August 27, 2026. ([University of Toronto][1])

# # # 🧬 The problem is surprisingly simple.

Some genetic mutations are called **nonsense mutations**.

Instead of producing the complete instructions for a protein, the mutation creates a premature STOP signal.

The ribosome reaches that signal and essentially says:

**“Stop. We're done.”**

The result?

The protein is cut short—or never properly produced at all.

Nonsense mutations account for an estimated **11% of inherited genetic disorders**, representing thousands of diseases, including certain forms of cystic fibrosis and neurological and muscular disorders. ([University of Toronto][1])

# # # Then comes the interesting part.

The Toronto team engineered tRNA molecules that can effectively **read through these premature stop signals**, allowing the cell to continue producing the full-length protein.

In cystic fibrosis models, the approach restored production of functional CFTR protein.

Even more interestingly, the researchers found that the strategy could work together with existing cystic fibrosis drugs.

In patient-derived organoids carrying nonsense mutations, the combination provided a response where either approach alone was insufficient. ([EurekAlert!][2])

And there is a much bigger idea behind this.

There are thousands of different disease-causing mutations.

But premature STOP signals come from only **three possible stop codons**.

That means one engineered therapeutic strategy could potentially be adapted to target the **same type of mutation across completely different genes and diseases.** ([University of Toronto][1])

This is very different from developing one drug for one mutation.

It suggests a future where we don't necessarily have to fix every genetic mistake individually.

Instead, we could develop **platform technologies that teach cells how to work around entire classes of genetic errors.**

Of course, this is still **preclinical research**, not an approved treatment.

The next major challenge is delivery: getting the right tRNA into the right tissues safely and efficiently. The researchers are already exploring specialized delivery systems, including particles that can potentially be aerosolized for lung applications. ([EurekAlert!][2])

But the bigger question is fascinating:

**Could RNA therapeutics eventually become programmable “repair tools” for entire categories of genetic disease—not by rewriting DNA, but by changing how cells interpret it?**

.alsesai.com

01/09/2026

Everyone is watching AI design new drugs.

But something more fundamental is happening underneath:

**AI is starting to learn how to make biological products more efficiently.**

RNA therapeutics are a perfect example.

Researchers are now combining large-scale experimental screening with deep learning to understand which DNA/RNA sequence designs can produce more RNA inside cells.

Instead of testing thousands of sequences one by one, scientists can:

**Design → Experiment → Measure → Train AI → Predict → Design again**

In one recent study, researchers experimentally tested around **100,000 sequence variants** and trained a deep-learning model to predict RNA production. The model achieved a remarkably high correlation with experimental results on independent data.

That changes the workflow.

Traditional biology often looks like:

> **Make something → test it → learn from failure.**

AI-assisted biology increasingly looks like:

> **Generate millions of possibilities → predict the best ones → test only the most promising candidates.**

And this could matter far beyond RNA.

Imagine AI eventually optimizing not only:

🧬 The therapeutic sequence
🏭 The biological production system
📦 The delivery mechanism
🧪 The manufacturing conditions
🤖 The entire experimental workflow

At that point, AI is no longer simply helping scientists **discover molecules**.

It is helping scientists **engineer biological production systems**.

That could become a major shift for mRNA vaccines, RNA therapeutics, gene therapy, and synthetic biology.

The bigger idea is even more interesting:

> **If biology is programmable, then eventually the entire process of designing AND manufacturing biology may become programmable.**

We may be moving from:

**AI-assisted biology**

to

# # # **AI-engineered biology.**

And the biggest breakthrough may not be the molecule AI discovers.

It may be the **biological factory AI learns to build.**

.alsesai.com

31/08/2026

CAR-T therapy has already changed cancer treatment.

But its next revolution may be even more radical:

**What if we didn't have to take your T cells out of your body at all?**

Today, conventional CAR-T therapy generally follows a complicated process:

**Patient → T cells extracted → genetic engineering → cell expansion → quality control → infusion**

It can take time, requires sophisticated manufacturing facilities, and is highly individualized.

Now researchers are exploring a fundamentally different approach:

# # # **Build CAR-T cells inside the patient.**

A recent study published in *Nature Materials* demonstrated a polymer–lipid nanoparticle system capable of delivering **CAR-encoding mRNA directly to T cells inside the body**.

Instead of removing T cells and engineering them in a laboratory, the researchers essentially deliver the **instructions** needed to turn existing T cells into CAR-T cells.

Think of it as the difference between:

> **Manufacturing the therapeutic cells outside the body**

and

> **Turning the patient's own cells into the therapy inside the body.**

In animal models, the approach generated functional CAR-T cells in vivo and produced therapeutic effects in models of cancer and fibrosis.

If this concept can eventually be translated safely to humans, it could change the economics and logistics of cell therapy.

Potentially:

⚡ Faster treatment
🏭 Less dependence on centralized cell manufacturing
💉 Simpler administration
🧬 More scalable personalized therapy
🌍 Greater access to advanced cell therapies

But there is a huge gap between an exciting animal study and an approved human treatment.

Researchers still need to solve critical problems including **targeting accuracy, dosage, durability, off-target effects, immune responses, and safety in humans**.

Still, the conceptual shift is remarkable.

We once thought of cell therapy as:

**Take cells → modify cells → give cells back.**

The next generation may look more like:

**Deliver instructions → reprogram cells → let the body manufacture its own therapy.**

And that raises a fascinating question:

# # # **What if the future of cell therapy isn't manufacturing cells for patients — but teaching patients' own cells how to become the medicine?**

.alsesai.com

29/08/2026

A small bat can live for **more than 40 years**.

And that isn't even the most surprising part.

Some bats can tolerate viruses that are dangerous to humans, show remarkable resistance to cancer, and maintain their bodies for decades despite their tiny size.

So scientists are asking a fascinating question:

**What does a bat know about aging that we don't?**

A new study published in *Nature* examined the genomes and cellular biology of eight species of *Myotis* bats, including exceptionally long-lived species.

The researchers found evolutionary changes linked to biological processes involved in **DNA repair, cancer resistance, immunity, and longevity**.

That is particularly interesting because these traits are usually studied separately.

But perhaps nature doesn't treat them as separate problems.

Maybe the ability to live longer isn't simply about slowing down aging.

Maybe it is about building a body that is exceptionally good at:

🧬 Repairing DNA damage
🛡️ Controlling inflammation
🦠 Tolerating viral infections
🧫 Eliminating damaged cells
🎯 Preventing abnormal cells from becoming cancer

In other words:

> **Longevity may not be one gene.
> It may be an entire biological system.**

And bats could be giving us a rare opportunity to study that system.

We shouldn't jump to the conclusion that humans can simply copy these genes and become 40-year-old bats.

Biology doesn't work that way.

But evolution has already conducted millions of years of experiments for us.

Different species have evolved radically different solutions to the same problems:

**How do you survive?
How do you repair damage?
How do you avoid cancer?
How do you stay functional for decades?**

Perhaps the next breakthrough in human longevity won't come from inventing something completely new.

**Perhaps we'll find it by decoding what nature has already invented.**

🦇 **What if the next great longevity drug begins with a bat?**

.alsesai.com

28/08/2026

# 🧬 Cancer May Begin Before the First Cancer Cell Exists.

We usually think cancer begins when a cell acquires a dangerous mutation.

But what if that isn't the whole story?

A growing body of research suggests that **aging can gradually reshape the biological state of our cells long before a tumor becomes detectable.**

As cells divide over time, their epigenetic landscape can drift. DNA methylation patterns change. Gene regulation becomes less stable. And in some tissues, cells may gradually lose aspects of their normal biological identity.

Recent research on **epigenetic drift** suggests that these age-related changes may be associated with cancer risk — raising an intriguing possibility:

> **The road to cancer may begin years before the first tumor appears.**

Think about the sequence differently:

**Aging → cellular changes → epigenetic drift → loss of cellular identity → increased cancer risk**

This could fundamentally change how we think about cancer prevention.

Today, much of cancer screening asks:

**“Can we find the tumor early?”**

But what if the next generation of screening asks:

**“Can we detect the biological changes that make a tumor more likely to appear?”**

Imagine a future blood test or tissue analysis that could identify abnormal cellular aging or epigenetic changes **years before cancer becomes clinically visible**.

We are not there yet.

Epigenetic drift is not the same thing as cancer, and scientists still need to determine which changes actually drive cancer and which are simply consequences of aging.

But the question is fascinating:

# # # **If we could detect a cell becoming biologically abnormal before it becomes cancerous, could we intervene before cancer even begins?**

Maybe the future of oncology isn't only about **finding cancer earlier.**

Maybe it's about finding the **first biological steps toward cancer**.

.alsesai.com

Alsesai

04/10/2026

24/09/2026

09/09/2026

08/09/2026

05/09/2026

01/09/2026

31/08/2026

29/08/2026

28/08/2026

Address

Website

Alerts

Shortcuts

Share

Category

Listing takedown request