The Animal Transcription Factor DataBase (AnimalTFDB) is a comprehensive TF database, in which we identified and classified all the genome-wide TFs in 50 sequenced animal genomes (Ensembl release version 60). In addition to TFs, the AnimalTFDB also collects transcription co-factors and chromatin remodeling factors of those genomes, which play regulatory roles in transcription. Here we defined the TFs as proteins containing a sequence-specific DNA-binding domain (DBD) and regulating target gene expression. Currently, the AnimalTFDB classifies all the animal TFs into 72 families according to their conserved DBDs.

• • 1. Methods for predicting TFs, transcription co-factors and chromatin remodeling factors
• • 2. TF families and seeds
• • 3. Gene annotation
• • 4. Web server

1. Methods for predicting TFs, transcription co-factors and chromatin remodeling factors:

The identification of TFs was based on the Hidden Markov Model (HMM) profiles of their DBDs. Among the 71 defined families, 59 families had HMM profiles for their DBDs in Pfam database (v25.0), while the remaining 12 families existing in InterPro database (v32.0) did not have available HMM profiles. To build the HMM profiles for them, we performed multiple sequence alignment by ClustalW2 for their DBD sequences and used the hmmbuild program in the HMMER package to build HMM profiles. Then, we applied the hmmsearch program to search the proteome sequences of each species to predict TFs. Based on our manual curation, we took E-value 0.0001 as the cutoff. Transcription co-factors are considered as proteins that interact with TFs in the transcription complex but do not bind to the DNA directly. The chromatin remodeling factors were defined as proteins that regulate transcription by modifying the chromatin formation. To identify them, we firstly got the human transcription co-factors and chromatin remodeling factors from TFCONES and GO database according to the items: transcription cofactor activity and chromatin remodeling, respectively. Then, we used the human sequences to perform BLAST search and chose the best BLAST hit as the transcription co-factor or chromatin remodeling factor for the searched species.

2. TF families and seeds:

Family DNA-binding domain Pfam ID or InterPro ID AF-4 AF-4 PF05110 AP-2 AP_2 IPR004979 ARID ARID PF01388 bHLH HLH PF00010 bZIP TF_bZIP bZIP IPR004827 C/EBP bZIP IPR016468 CBF CBF_alpha PF02312 COE COE IPR003523 CSL Beta-trefoil PF09270 CG-1 CG-1 PF03859 CP2 CP2 PF04516 CSD CSD PF00313 E2F E2F_TDP PF02319 ETS Ets PF00178 Fork head Fork_head PF00250 GCM GCM PF03615 GTF2I GTF2I PF02946 HMG HMG_box PF00505 HMGI/HMGY HMGI/HMGY IPR000116 Homeobox Homeobox Homeobox PF00046 Pou Homeobox, Pou PF00157 CUT Homeobox, CUT PF02376 TF_Otx Homeobox, TF_Otx PF03529 HSF HSF_DNA-bind PF00447 HTH HTH_psq PF05225 IRF IRF PF00605 MH1 CTF/NFI MH1 PF00859 MH1 MH1 PF03165 MYB Myb_DNA-bd PF00249 MBD MBD PF01429 NDT80/PhoG NDT80_PhoG PF05224 NF-Y NF-YA CBFB_NFYA PF02045 NF-YBC CBFD_NFYB_HMF PF00808 Nrf1 Nrf1_DNA-bind PF10491 PC4 PC4 PF02229 P53 P53 PF00870 PAX PAX PF00292 Prox1 Prox1 PF05044 RFX RFX PF02257 RHD RHD PF00554 Runt Runt PF00853 SAND SAND PF01342 SRF SRF PF00319 STAT STAT_bind PF02864 T-box T-box PF00907 TEA TEA PF01285 TSC22 TSC22 PF01166 Tub Tub PF01167 Zinc finger zf-C2HC zf-C2HC PF01530 zf-GAGA zf-GAGA PF09237 zf-BED zf-BED PF02892 zf-C2H2 ZBTB zf-C2H2 PF00651 zf-C2H2 zf-C2H2 PF00096 Nuclear Receptor(zf-C4) PPAR receptor zf-C4 IPR003074 Androgen receptor zf-C4 PF02166 COUP_TF zf-C4 IPR003068 Ecdystd receptor zf-C4 IPR003069 GCR zf-C4 PF02155 Nuclear hormone receptor zf-C4 IPR003070 Oestrogen receptor zf-C4 PF02159 Progesterone receptor  zf-C4 PF02161 Retinoic acid receptor zf-C4 IPR003078 ROR receptor  zf-C4 IPR003079 Thyroid hormone receptor  zf-C4 IPR001728 Other nuclear receptor zf-C4 PF00104 DM DM PF00751 zf-GATA zf-GATA PF00320 zf-LITAF-like zf-LITAF-like PF10601 zf-MIZ zf-MIZ PF02891 zf-NF-X1 zf-NF-X1 PF01422 THAP THAP PF05485 Others

3. Gene annotation:

1) Gene basic information Gene basic information includes Ensembl ID, Gene ID, symbol, alias, full name, other designations, chromosome location and transcripts etc. All of the gene information was extracted from gene_info file downloaded from NCBI ftp (ftp://ftp.ncbi.nih.gov/gene/DATA/ ) and gtf files downloaded from Ensembl ftp (ftp://ftp.ensembl.org/pub/release-60/gtf/) 2) Gene structure The gene structure describes the distribution of CDS, UTR and intron of gene on chromosome. 3) Gene ontology (GO) annotation The GO annotations were parsed from gene2go file, which was downloaded from NCBI ftp ( ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/); 4) Function domain annotation The function domain displays the domains distribution of the longest protein of each gene. All of the protein domains were identified by HMMER profiles downloaded from Pfam database. 5) 3D structure hit 3D structure hit of each gene was gotten from PDB database. We applied BLAST program to search the protein sequences of factors against sequences in PDB with cutoffs: E-value<=1e-15, hit length>=50 amino acids and identity>=55%. 6) Protein-protein Interaction The protein-protein interactions were extracted from BioGRID and HPRD databases and a literature (PMID:20211142). 7) Pathway annotation The pathway annotations were obtained from the KEGG and BioCarta databases. 8) Ortholog information We used the reciprocal best blast hit (RBH) strategy to predict the ortholog, on the assumption that two genes in two genomes will find each other as the best hit in the other genome. 9) Paralog Information The BLAST Score Ratio (BSR) approach was used to identify the paralog. Paralogs are required BLAST identity>=50%, E-value<=1e-20, coverage>=70%, BSR>=0.4. 10) Cross Reference Cross reference provides links to other database including Unigene, UniSTS, Genebank, OMIM, HGNC, CGNC etc.

4. Web server:

Current URL of AnimalTFDB is http://www.bioguo.org/AnimalTFDB/. Users can browse or search the data at different levels. Browse: 1) Browse by species. Users can browse data by clicking the logo of species or by clicking the name on the left treeview; 2) Browse by families. Users can browse data by clicking the logo of families or by clicking the name on the left treeview. Search: 1) Quick search for Ensembl ID for gene, transcript and protein, Entrez gene ID or gene symbol at head of each page; 2) Advanced search page provides multiple ways to search. Users can search by different basic information and annotation information of a TF. Before search, users must choose the families and species.