The transcription regulatory code of a plant leaf

The transcription regulatory network underlying essential and complex functionalities inside a eukaryotic cell is defined by the combinatorial actions of transcription factors (TFs). However, TF binding studies in plants are too few in number to produce a general and comparative picture of this complex regulatory network. Here, we used ChIP-seq to determine the binding profiles of 104 TF expressed in the maize leaf (Data can be downloaded from NCBI SRA under accession number PRJNA518749) With this large dataset, we trained machine-learning models to identify TF sequence preferences. A contrast between Maize and Arabidopsis TF sequence preferences revealed that DNA binding follows the conservation of TF protein families. Finally, the trained models were used to predict and compare the regulatory networks in other grasses species (Sorghum and Rice), which revealed that the edges between TF and TF coding genes are more likely to be maintained (i.e., evolutionarily conserved). On a practical level, we expect the presented TF binding models to be integrated into pipelines to predict effects of non-coding variants, both common and rare, on TF binding, to pinpoint causal sites. As the possibility of being able to predict and generate novel variation not seen in nature could fundamentally change future plant breeding. Detail: Each *tar.gz file is a bag-of-k-mer model fitted for a single ZmTF, which can be used for predictions. Information about each ZmTF is included in the table tfids.tsv For more information about the project: The transcription regulatory code of a plant leafThis work was funded by USDA-ARS, NSF PGRP #1238014, National Science Foundation of China 31871313, National Key Research and Development Program of China 2016YFD0101003, Taishan Pandeng program as well as Hong Kong UGC GRF 14108117, Area of Excellence Scheme (AoE/M-403/16)