Prioritization of Androgenetic Alopecia-Associated Lactylation Sites through Multi-Omics Integration and the HybridKla Deep Learning Framework
This study integrates multi-omics data from scalp tissues with a HybridKla deep learning framework to prioritize androgenetic alopecia-associated lactylation sites, specifically identifying ZBTB7B K206 and K212 as key candidates for investigating post-translational regulation in the follicular immune microenvironment.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
Hair loss is often seen as a cosmetic issue, but for the millions of people affected by androgenetic alopecia, it is a complex biological puzzle. The condition, commonly known as male or female pattern baldness, involves hair follicles shrinking over time until they can no longer produce visible hair. While scientists have long studied the genes and proteins involved in this process, a newer layer of biology has recently come into focus. Cells do not just rely on the instructions written in their DNA; they also use chemical tags to attach to their proteins, acting like switches that turn specific functions on or off. One such tag, called lactylation, links the cell's energy metabolism to how it behaves. When a cell produces lactate, a byproduct of energy use, it can attach this molecule to proteins, potentially changing how those proteins work. Understanding these tiny chemical changes could reveal why hair follicles stop growing in certain areas of the scalp, offering a new way to look at the disease.
A team of researchers at Tongji University in Shanghai decided to investigate this possibility by looking at the scalp itself. They gathered tissue samples from the front and back of the heads of several men with hair loss. The front of the head, where hair typically thins, was compared to the back, where hair usually remains thick. The scientists used advanced tools to read the genetic activity of individual cells, measure the total amount of proteins present, and specifically hunt for those proteins carrying the lactylation tag. Their goal was not to find a cure immediately, but to identify specific molecular clues that could explain what is happening inside the follicle. They focused on a protein called ZBTB7B, which is known to help regulate immune cells, because previous studies suggested that immune activity plays a role in hair loss.
The researchers found a striking contradiction in their data. In the thinning hair on the front of the scalp, the total amount of the ZBTB7B protein was actually higher than in the thick hair at the back. However, when they looked specifically for the lactylation tags attached to this protein, they found the opposite trend. The specific spots on the protein where these tags usually attach were much less common in the thinning hair. It was as if the factory was producing more of the machine, but fewer of the machines had the specific tool attached that might change how they work. This pattern was consistent across the different donors, suggesting it was a real biological signal rather than a random error. The two specific spots on the protein where this difference was most noticeable were identified as K206 and K212.
To make sense of these findings, the team turned to a computer model designed to predict where these lactylation tags are likely to appear on a protein. They adapted a powerful artificial intelligence framework, originally built to recognize patterns in protein sequences, to analyze their specific data. By training the model on thousands of known examples and then testing it on the proteins from their study, they confirmed that the spots K206 and K212 were indeed likely candidates for lactylation. The computer model, combined with the physical evidence from the lab, pointed to these two locations as the most promising places to look next. The researchers also checked the structure of the protein using a digital map of its shape, noting that these spots sit in a flexible region between two major functional parts of the molecule, making them plausible sites for regulation.
The study did not prove that lactylation causes hair loss, nor did it show that fixing these tags would regrow hair. Instead, it provided a clear, prioritized list of targets for future experiments. The work suggests that the immune system's role in hair loss might be controlled by these tiny chemical switches on a specific protein. By narrowing the focus from the entire genome or proteome down to just two specific spots on one protein, the researchers have given the scientific community a concrete starting point. They have shown that looking at the total amount of a protein is not enough; one must also look at how that protein is chemically modified. This approach connects the broad environment of the scalp tissue with the precise molecular changes happening inside individual cells, offering a more detailed map for understanding the biology of hair loss.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.