Modeling 0.6 million genes for the rational design of functional cis-regulatory variants and de novo design of cis-regulatory sequences

• Published in: Proceedings of the National Academy of Sciences - PNAS Vol. 121; no. 26; p. e2319811121
• Main Authors: Li, Tianyi, Xu, Hui, Teng, Shouzhen, Suo, Mingrui, Bahitwa, Revocatus, Xu, Mingchi, Qian, Yiheng, Ramstein, Guillaume P., Song, Baoxing, Buckler, Edward S., Wang, Hai
• Format: Journal Article
• Language: English
• Published: United States National Academy of Sciences 25.06.2024
• Subjects: Algorithms; Annotations; Binding sites; Binding Sites - genetics; Biological Sciences; Biophysics and Computational Biology; Conserved sequence; Deep Learning; Deoxyribonucleic acid; Design; DNA; Gene Expression Regulation, Plant; Gene sequencing; Genes, Plant; Genetics; Models, Genetic; Nucleotide sequence; Nucleotides; Physical Sciences; Plant Biology; Plant species; Plants - genetics; Polymorphism, Single Nucleotide; Population genetics; Quantitative Trait Loci; Regulatory sequences; Regulatory Sequences, Nucleic Acid - genetics; Single-nucleotide polymorphism; Transcription Factors - genetics; Transcription Factors - metabolism; Zea mays - genetics; deep learning; synthetic biology; engineering; transcriptional regulation; cis-engineering
• ISSN: 0027-8424; 1091-6490; 1091-6490
• DOI: 10.1073/pnas.2319811121

SignificanceThe enormous variation space and obscure syntax rules of eukaryotic transcriptional regulatory DNA sequences hamper their rational design. Here, we developed PhytoExpr, a deep learning framework that reads regulatory DNA sequences to predict their messenger ribonucleic acid (mRNA) abundance and also the plant species they are from. PhytoExpr was trained over major clades of the plant kingdom to make predictions on unseen gene families from unseen species. The sequence features learned by PhytoExpr were enriched with conserved noncoding sequences, transcription factor binding sites, and eQTLs. We also fit PhytoExpr into two algorithms for the rational design of functional cis-regulatory variants for genome editing, as well as the de novo design of species-specific cis-regulatory DNA sequences for synthetic biology. Rational design of plant cis-regulatory DNA sequences without expert intervention or prior domain knowledge is still a daunting task. Here, we developed PhytoExpr, a deep learning framework capable of predicting both mRNA abundance and plant species using the proximal regulatory sequence as the sole input. PhytoExpr was trained over 17 species representative of major clades of the plant kingdom to enhance its generalizability. Via input perturbation, quantitative functional annotation of the input sequence was achieved at single-nucleotide resolution, revealing an abundance of predicted high-impact nucleotides in conserved noncoding sequences and transcription factor binding sites. Evaluation of maize HapMap3 single-nucleotide polymorphisms (SNPs) by PhytoExpr demonstrates an enrichment of predicted high-impact SNPs in cis-eQTL. Additionally, we provided two algorithms that harnessed the power of PhytoExpr in designing functional cis-regulatory variants, and de novo creation of species-specific cis-regulatory sequences through in silico evolution of random DNA sequences. Our model represents a general and robust approach for functional variant discovery in population genetics and rational design of regulatory sequences for genome editing and synthetic biology.