PlantDTI Help and Documentation


Download User Manual


The Plant-DTI was implemented as a user-friendly application tool to predict the interactions between DBD and TFBS in plants. It will facilitate all interested users to fill the knowledge gap of TF-TFBS interactions or TF-Target gene interactions and improve understanding of transcriptional regulation in plant species. This tool was constructed based on machine learning approach using experimental data of DNA binding domain (DBD) and transcription factor binding site (TFBS) interactions from CIS-BP version 1.02 database (1).

Currently, the Plant-DTI can support only monotypic transcription factors, TFs with single DBD, and cover 26 DBD types of 30 TF families in plants. The available DBD types and TF families in Plant-DTI are listed in tutorial section.


How to use tool?

Running a PlantDTI prediction from the online web application tool is simple, users can try with the example sequence.

1. Users set the input sequences
2. Users set the tool parameter
3. Users submit the query input sequence

In the page of Plant-DTI online tool, you can click on each button. The meaning of each button is described below.

An input session of the model prediction
represents the definition of each input data of Plant-DTI.
(2) Example sequence
The example of input sequences including amino acid sequences of DBD of TF protein, nucleotide sequence of each promoter of target gene.
(3) Clear sequence
Delete query inputs.
(4) Submit
Submit the query inputs to predict the DBD-TFBS interaction.

Input data for Plant-DTI

Plant-DTI required 3 inputs for predicting DBD-TFBS interaction.

1. Amino acid sequence of DNA binding domain (DBD)

The amino acid sequence of DBD which is the region in TF protein binding to DNA. In the online web application tool, user can use amino acid sequence of transcription factor or DNA binding domain as an input. The tool will identify its DBD sequence and DBD type within the sequence using Pfam scan (2). There are 26 DBD types of monotypic TFs (single-DBD TFs) available in Plant-DTI as shown in table below. The information of TF family and DBD type was based on the literatures (1, 3).

TF family DBD type Pfam ID
ERF AP2 PF00847
AT hook AT_hook PF02178
B3 B3 PF02362
bZIP bZIP_1 PF00170
CAMTA CG-1 PF03859
LBD DUF260 (LOB) PF03195
GeBP DUF573 PF04504
BES1 DUF822 (BES1_N) PF05687
E2F/DP E2F_TDP PF02319
EIL EIN3 PF04873
FAR1 FAR1 PF03101
bHLH HLH PF00010
Sox HMG_box PF00505
HB-other Homeobox (Homeodomain) PF00046
HB-PHD Homeobox (Homeodomain) PF00046
HD-ZIP Homeobox (Homeodomain) PF00046
TALE Homeobox (Homeodomain) PF00046
WOX Homeobox (Homeodomain) PF00046
MADF MADF_DNA_bdg* PF10545*
MYB_related Myb_DNA-binding PF00249
C2H2 zf-C2H2** PF00096**
Dof zf-Dof PF02701
* MADF_DNA_bdg (PF10545) is similar to Myb_DNA-bind4 (PF13837) (2)
** zf-C2H2 (PF00096) is similar to zf-C2H2_6 (PF13912) (2)
2. Promoter sequence

DNA sequences of a promoter region of an interested gene.

3. Interaction probability threshold

The criteria for predicting interaction between query DBD and TFBS sequence in Plant-DTI. The suggested value for prediction is in a range of more than 0.5 to 1. Higher interaction probability threshold represents the higher confidence of predicted interaction. The default threshold is 0.7.

Results of Plant-DTI

The predicted result of Plant-DTI is shown in the figure including, the region of DNA-binding domain in the amino acid sequence query and predicted TFBS sequences in the interested promoter query.

(1) DNA-binding domain sequence
Amino acid sequences of DNA-binding domain in a TF protein. It was highlighted in red and shown its DBD type based on Pfam identification.
(2) Transcription factor binding sites on promoter sequence
DNA sequences of a promoter region of an interested gene. The predicted TF binding sites on both strands (+ strand and - strand) were highlighted in blue.

The predicted interaction results were shown in table format. User can sort and filter the result based on any keywords in the column e.g., prediction score equal to 1 for obtaining the highest confident predicted interactions. User can download the table result in csv format via Export CSV button. The meaning of each column result is described below.

TF binding site
Predicted TF binding site on the queried promoter sequence.
Predicted TF binding site sequence length.
The start position of predicted TF binding site, the reported result relied on position of positive strand used as input sequence.
The end position of predicted TF binding site, the reported result relied on position of positive strand used as input sequence.
Strand of DNA which TF binding site is located.
Model type
Method to generate negative data for Plant-DTI tool which can be either random DBD-TFBS pairs (RP) or random nucleotide within TFBS sequence (RW).
Prediction score
The interaction probability predicted via Plant-DTI for each queried pair between DBD and TFBS.

1. Weirauch,M.T., Yang,A., Albu,M., Cote,A.G., Montenegro-Montero,A., Drewe,P., Najafabadi,H.S., Lambert,S.A., Mann,I., Cook,K., et al. (2014) Determination and Inference of Eukaryotic Transcription Factor Sequence Specificity. Cell, 158, 1431–1443.

2. Mistry,J., Chuguransky,S., Williams,L., Qureshi,M., Salazar,G.A., Sonnhammer,E.L.L., Tosatto,S.C.E., Paladin,L., Raj,S., Richardson,L.J., et al. (2021) Pfam : The protein families database in 2021. Nucleic Acids Res., 49, D412–D419.

3. Jin,J., Zhang,H., Kong,L., Gao,G. and Luo,J. (2014) PlantTFDB 3.0: a portal for the functional and evolutionary study of plant transcription factors. Nucleic Acids Res., 42, D1182–D1187.


Ruengsrichaiya B, Nukoolkit C, Kalapanulak S and Saithong T (2022) Plant-DTI: Extending the landscape of TF protein and DNA interaction in plants by a machine learning-based approach. Front. Plant Sci. 13:970018. doi: 10.3389/fpls.2022.970018