A parallel convolution attention network for crop genomic prediction

Wenyu Peng, Yuxia Sheng, Ying Zhou · 2026

Read A parallel convolution attention network for crop genomic prediction by Wenyu Peng, Yuxia Sheng, Ying Zhou (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0358451.txt.

Description

Wenyu Peng, Yuxia Sheng, and Ying Zhou present a deep learning method for predicting crop traits from DNA, in PLOS ONE in 2026. Breeders increasingly select plants by genome rather than waiting to grow them. But the bond between markers and traits bends in ways classical statistics handle poorly. Earlier neural networks also struggled, with vanishing signals and heavy computation. Their network runs convolutional filters in parallel rather than in sequence, and adds a coordinate attention step that weighs marker channels and positions jointly.

On four benchmark datasets, the method outperformed eight established rivals, spanning a traditional statistical model, three machine learning approaches, and four popular deep learning systems. The authors report the accuracy gains as significant across the board. For breeding programs, better genomic prediction means fewer field seasons wasted on plants whose genes already betray mediocre yield, quality, or resilience.

Catalog record

CREATORWenyu Peng, Yuxia Sheng, Ying Zhou
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40624
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0358451

Browse the full library · About the Tower of Babel