For beginners

Browse

  1. Browse by species. In the AnimalTFDB database, we separated all 183 species into 16 categories referencing the ensemble database. In addition, to make it convenient for users to browse, we put a favourites category on the top of the species page. Users can browse the database by clicking the logo of species, species->families->family gene list->single gene annotation is applied for this browse way. The family gene list page also shows the multiple sequence alignment of the DBDs, the weblogo graph of the multi-alignment of these TFs, and a brief introduction and reference of this TF family.
  2. Browse by family. We separated TFs into 73 families and TF Cofactors into 82 families. Furthermore, 73 TF families and 83 TF Cofactor families were classed into 6 and 6 categories separately. Users can browse the database by clicking the logo of TF families or the bars of transcription cofactors. The cascading style of families->species->family gene list->single gene annotation is applied for this browse way.

Search page provides multiple ways to search the database. Users can search by different basic information (multiple gene ids, gene symbol, alias and full name), mRNA or protein expression of a TF. For the expression search, specific species, types of tissues, cell lines, development stages and cancers, and the lowest threshold of gene expression levels could be selected to filter the search result. In version 4.0, we have also added the function to batch search, and Users can search a list of genes by different basic information to determine whether the gene is in our database.

TF Predicition

We set up a TF prediction server to help users identify TFs from their protein sequences. The prediction method and TF family assignment rules can be found in the prediction and TF family assignment rules pages. Currently, users can upload up to 1000 protein sequences at one time and obtain results within a few minutes. In the prediction result, TF family, alignment e-value, and detailed alignment information will be provided.

TF binding site predicition

To help users identify TF targets from their nucleotide sequences, a TFBS prediction server was provided. We collected the TF motif matrixes from TRANSFAC, JASPAR, HOCOMOCO, CIS-BP, and hTFtarget databases. We used these matrixes to scan the alignment sequence of human to predict the TF targets by using motif detection function of The MEME Suite (http://meme-suite.org/index.html). The prediction result shows the TF binding site sequence, score, P-value, Q-value, and detailed alignment information. Furthermore, specific TF results can be shown through the filter bar. The calculation of score, P-value, and Q-value of prediction result can be seen in the article (PMID: 21330290). The online help documentation of FIMO tool is available here.

If users want to predict TFBS locally, the user can first download the TFBS prediction files in the Download interface, then run FIMO and use fourdatabse_all.meme and hTFtarget_prediction.motifs_matrix.meme to get the prediction results respectively, and finally use fourdatabse_human_mouse.annotation and hTFtarget.annotation to annotate the prediction results (using the ids in the meme file to match and thus obtain the TFBS prediction results in human and mouse).

BLAST

To help users find homologous genes and explore functions of poorly studied TFs, a BLAST tool was provided to search against TFs in our database with protein or DNA sequences. The protein sequences of all species or one specific species could be selected for the BLAST database. The specific e-value can be chosen.

Download

The gene lists and protein sequences of TFs and transcription cofactors for each species could be downloaded from the download page.