PlantPAN3.0: a new and updated resource for reconstructing transcriptional regulatory networks from ChIP-seq experiments in plants

Abstract

The Plant Promoter Analysis Navigator (PlantPAN; http://PlantPAN.itps.ncku.edu.tw/) is an effective resource for predicting regulatory elements and reconstructing transcriptional regulatory networks for plant genes. In this release (PlantPAN 3.0), 17 230 TFs were collected from 78 plant species. To explore regulatory landscapes, genomic locations of TFBSs have been captured from 662 public ChIP-seq samples using standard data processing. A total of 1 233 999 regulatory linkages were identified from 99 regulatory factors (TFs, histones and other DNA-binding proteins) and their target genes across seven species. Additionally, this new version added 2449 matrices extracted from ChIP-seq peaks for cis-regulatory element prediction. In addition to integrated ChIP-seq data, four major improvements were provided for more comprehensive information of TF binding events, including (i) 1107 experimentally verified TF matrices from the literature, (ii) gene regulation network comparison between two species, (iii) 3D structures of TFs and TF-DNA complexes and (iv) condition-specific co-expression networks of TFs and their target genes extended to four species. The PlantPAN 3.0 can not only be efficiently used to investigate critical cis- and trans-regulatory elements in plant promoters, but also to reconstruct high-confidence relationships among TF-targets under specific conditions.

Figure 1.

The output interfaces of ‘Protein Search’ in PCBase in PlantPAN 3.0. (A) After users select a species and a regulatory factor (marked in red boxes), detailed information for the selected dataset (right) is displayed. The result page also provides (B) a searchable table for Target Browse, where user can click ‘Visualize’ to identify the location of a binding site, (C) binding proportion, (D) motif logos, (E) Peak Browse for a dataset and (F) tables to download processed files and link with external databases.

Figure 2.

The binding sites for SEP3, AP1, H3K4me3, FLC, and AGL15 across SOC1 gene (AT2G45660) in the Jbrowse viewer. The upstream region (+500 to -2000; Chr2:18813047–18810548) of SOC1 is highlighted with a yellow background. The experimental binding sites from ChIP-seq are shown in the first three tracks. The last three lines are the predicted TFBSs via PWM patterns.

Figure 3.

Conserved transcription regulation between YUC8 and its homolog (Glyma.03G169600). (A) The partial TFBS prediction results in the seventh conserved regions, where RVE1 and four homeodomain-like superfamily proteins were marked in red. (B) The transcriptional regulatory network of YUC8 (pink node) and Glyma.03G169600 (light yellow node). Red and yellow nodes represent predicted TFs from A. thaliana and G. max, respectively. Green line were used to link homologous TFs.