Global EditionASIA 中文双语Français
Life

DNA, AI raise martyrs from ashes

University course merges technology with archaeology, giving life to those who made the ultimate sacrifice during the Red Army's historic Long March, Wang Xin reports in Shanghai.

By Wang Xin | CHINA DAILY | Updated: 2026-04-11 00:00
Share
Share - WeChat
Researchers from the Molecular Archaeology Lab at Fudan University pay tribute to Red Army martyrs in Zunyi, Guizhou province. The team has completed DNA identifications for 12 martyrs in the city. CHINA DAILY

Deng Ping (1908-35), a senior commander who made great sacrifices in the fight for victory more than 90 years ago in Zunyi, Guizhou province, recently rose again to greet the world with a smile.

By merging artificial intelligence with archaeology, students and researchers at Fudan University in Shanghai restored the image of the famous 27-year-old martyr.

Deng was one of the highest-ranking commanders who sacrificed himself in the historic Long March (1934-36), which marks the 90th anniversary of its victory this year.

The Long March was the epic retreat by the Red Army of the Communist Party of China and a defining event in Chinese history. Filled with enduring spirit, it is an important chapter in the Chinese revolution's journey from setbacks to victory.

"As a senior Red Army commander who fell during the Long March, Deng Ping has no clear portrait that has survived to the present day, which the staff members at the Zunyi Red Army Martyrs Cemetery greatly regret," says Wen Shaoqing, an associate professor from the Molecular Archaeology Lab at Fudan University.

For many years, people could only imagine Deng's story through a rough black-and-white sketch drawn from his comrades' memories.

With the wider application of AI technology, the university officially launched the "AI Archaeology" course in the fall semester of 2024. The course aims to explore innovative intersections between the two disciplines, and students are encouraged to create various projects based on their interests, applying AI solutions to the field of archaeology.

The restoration of Deng's image is one result. Through repeated trial and error, participants have made advancements in various areas, including collecting historical materials, conducting research, AI generation and refinement, scene reconstruction, and more.

"AI does not create a face that never existed. Instead, it generates a historically accurate, period-appropriate visage based on limited historical imagery and documentation," Wen explains.

When the smiling Deng finally came to life from the history archives, Wang Weijia, a postgraduate student involved in the project, felt deeply moved.

"Technology need not be profit-driven to possess serious social value. The martyrs' reconstruction project has no business model, but it gives technology tangible meaning," says Wang.

Deng is among the few whose names are preserved in historical records. In Zunyi, over 3,000 Red Army soldiers gave their lives in tough battles during the Long March, the majority of whom remain unknown heroes.

Wen's team has been utilizing modern molecular archaeology techniques since 2015 to find their relatives. In 2023, the team expanded their research to comprehensively uncover the martyrs' names, images, family members, physical conditions, and life records.

To date, the team has completed DNA identifications for about 1,600 martyrs across China, restored the appearance of over 60 heroes, and helped more than 10 families of martyrs find their relatives.

Despite previous success, the team found it challenging when it received a mission from the Zunyi Red Army Martyrs Cemetery in 2024 to conduct DNA testing on the remains of 16 martyrs from the city and help identify their relatives.

The team explains that these martyrs' remains were doused in turpentine, then cremated. High temperatures can cause DNA to break into very short fragments or even degrade completely. The shorter the DNA fragments, the more difficult they are to extract and sequence.

To tackle the challenge, Wen's team optimized various aspects of their original solutions to specifically target the shorter DNA fragments and obtain the DNA data used for identification. Eventually, they obtained the DNA profiles from 14 remains. They also found that three of the 16 remains were from one individual, enabling them to identify the families of 12 martyrs from Zunyi.

"The DNA of these martyrs is often severely degraded, highly damaged, and heavily contaminated. Also, many of them died very young with no children. After 90 years, their distant relatives are three to five generations removed," Wen says.

He further explains that the nation's existing public security DNA database houses much longer DNA fragments and therefore does not align with the data type of these martyrs. The team says it needs to establish a new database featuring more and shorter genetic markers for more comprehensive and intricate identification and analysis, at a cost several times higher.

"We hope to bring light to the martyrs, as they bravely gave their lives for a brighter future for us, to make the unknown known, and to help them return home to their families. We also hope more people can learn about them, their stories and their spirits, and carry them forward," says Wen.

The Qingsong Hall at Zunyi Red Army Martyrs Cemetery in Guizhou province. CHINA DAILY
The Zunyi Red Army Martyrs Cemetery. CHINA DAILY
An AI-restored photo (left) and a sketch (right) of Deng Ping. CHINA DAILY
An AI-restored photo (left) and a sketch (right) of Deng Ping. CHINA DAILY

Today's Top News

Editor's picks

Most Viewed

Top
BACK TO THE TOP
English
Copyright 1994 - . All rights reserved. The content (including but not limited to text, photo, multimedia information, etc) published in this site belongs to China Daily Information Co (CDIC). Without written authorization from CDIC, such content shall not be republished or used in any form. Note: Browsers with 1024*768 or higher resolution are suggested for this site.
License for publishing multimedia online 0108263

Registration Number: 130349
FOLLOW US