Jingjing Zhai

Convergent genome- and gene-level constraints shape repeated environmental adaptation in grasses

This study identifies convergent genomic and gene-level constraints in grasses adapting to extreme environments using phylogenomic data and large language models.

The maize recombination landscape evolved during domestication.

This study shows that maize evolved a higher recombination rate and re-patterning of crossover events during domestication, which enhanced genetic diversity and adaptation.

PlantCAD2: a DNA foundation model for interpreting genomes across flowering plants

PlantCAD2 is a DNA foundation model pre-trained on 65 angiosperm genomes, showing improved performance in evolutionary conservation and chromatin accessible region prediction compared to existing models.

A transcriptomic atlas of grass senescence reveals divergent underground sink networks limit nitrogen recycling in annuals

The paper constructs a transcriptomic atlas of grass senescence to identify molecular mechanisms of nitrogen recycling, finding divergent underground sink networks in annual crops.

Extensive modulation of a conserved cis-regulatory code across 589 grass species

The paper identifies a conserved cis-regulatory code across 589 grass species and finds that motif conservation declines rapidly but remains significant over evolutionary time, with adaptive changes linked to ecological transitions.

GeneCAD: Plant Genome Annotation with a DNA Foundation Model

GeneCAD improves gene annotation in plant genomes using a DNA foundation model, achieving higher accuracy than existing tools across diverse species.

Fishing for a reelGene: evaluating gene models with evolution and machine learning

reelGene evaluates gene models using machine learning and evolutionary data, finding 28% of maize gene models are incorrectly annotated or nonfunctional.

deepTFBS: Improving within- and cross-species prediction of transcription factor binding using deep multi-task and transfer learning

deepTFBS improves TFBS prediction within and across species using deep multi-task and transfer learning, outperforming existing methods in Arabidopsis and wheat.

Constrained evolution of a core winter proteome across independently cold-adapted PACMAD grasses

The paper compares cold-responsive proteomes across five PACMAD grasses to assess conservation of protein-level responses and identifies LEA3 as a key cold-tolerant protein.

Extensive genome evolution distinguishes maize within a stable tribe of grasses

The paper examines genome evolution in maize and related grasses, revealing variable outcomes of polyploidy across species.

Cross-species modeling of plant genomes at single nucleotide resolution using a pre-trained DNA language model

The paper introduces PlantCaduceus, a DNA language model for plants, showing high transferability across species and improved performance in mutation prediction and variant identification.

easyMF: A Web Platform for Matrix Factorization-based Biological Discovery from Large-scale Transcriptome Data

easyMF is a web platform that uses matrix factorization to analyze large-scale transcriptome data, enabling biological discovery through metagene-based analysis.

DeepGS: Predicting phenotypes from genotypes using Deep Learning

DeepGS predicts phenotypes from genotypes using deep learning, achieving higher accuracy than conventional methods and RR-BLUP across multiple traits.

PEA: an integrated R toolkit for plant epitranscriptome analysis

PEA is an R toolkit for plant epitranscriptome analysis, offering CMR calling, prediction, and annotation with high accuracy for m6A modifications in Arabidopsis thaliana.