Inside the Jingdezhen Imperial Kiln Institute, rows upon rows of gray metal cabinets house thousands of transparent drawers filled with ancient ceramic shards. Together, they form what the institute describes as the world's first ancient ceramic "gene bank" — a database that stores the "DNA" of broken pottery.
Jingdezhen city in East China, known as the "porcelain capital", has produced ceramics for over 1,700 years. Beneath the city's famed masterpieces lie layers of broken fragments several meters thick — flawed imperial ceramics and damaged folk products. To date, nearly 20 million shards have been unearthed from the city's deep archaeological deposits.
The Jingdezhen Imperial Kiln Institute, together with Tsinghua University, the Shanghai Institute of Ceramics under the Chinese Academy of Sciences, and other partners, has built a ceramic "gene bank" to collect data points from the shards.
The bank offers a standardized physical sample library and a unified data framework to support international authentication and academic research. Its data encompasses shape, decoration, body material, glaze, color, firing marks and inscriptions.
One million records
To become a "genetic" sample, each shard is first cleaned and cut into four physical forms — fragments, cross-sections, thin slices and powder. Scanning electron microscopy is used to examine these prepared samples.
Extracting a gene means identifying key characteristics, explains Tong Yuting, a staff member at the Jingdezhen Imperial Kiln Institute. "For blue and white porcelain, we don't just look at the perfectly colored pieces. We also collect shards with paler tones, deeper shades, or even rust spots, as all of these variations are considered feature points."
By gathering this full spectrum of color variations, researchers build a complete "color chart" for blue-and-white wares. Analyzing these variations enables researchers to reverse-engineer the cobalt pigment composition, formula ratios, and firing techniques of ancient craftsmen, making each shard a window into historical production methods.
The database now holds over 1 million genetic records from 12,000 specimens across 3,000 sets, supported by a standardized system for sample processing, management and preservation. This establishes a benchmark system for dating and authentication.
Maria Mayer, curator at the Medeiros e Almeida Museum in Portugal, is studying a group of blue-and-white porcelain pieces fired in the Ming Dynasty (1368-1644) bearing Portuguese coats of arms. "The gene bank offers a scientific and systematic foundation for classifying ceramic materials, which will help me uncover the cultural exchange history behind this collection," she says.
Xin Wenyuan, a researcher at the British Museum, notes that many overseas collections have traditionally been dated and classified solely by style. "The archaeological sequence and technical data from the imperial kiln site give us a more reliable basis for comparison," she says. With the gene bank, Western ceramic studies can move from "appreciating finished products" to analyzing materials, technology and production systems.
Xin adds that Jingdezhen's gene bank lets researchers observe different stages of production, including glaze composition, trimming methods, firing temperature control and even disposal.
"I noticed a branch-and-flower pattern on a blue-and-white sauce-glazed bowl fragment that closely resembles pieces in our collection and those from the 19th-century Desaru shipwreck. It's like a matching game, one decoration connects the whole chain — production, transport, consumption, and museum collection."
The data also helps with restoration. Researchers recently reassembled two large dragon-decorated vats made in the Ming Dynasty from 16,000 fragments — a task once thought impossible. AI algorithms processed each shard's genetic data to predict correct assembly paths, bringing the relics back to life.
