TFBSTools

Dependencies & Environment

Package-intrinsic requirements from the Bioconductor landing page, reproduce in any R environment.

• Version: 1.50.0 · Bioconductor: 3.23 · R: ≥ 4.6
• Imports: Biobase, Biostrings, pwalign, BiocGenerics, BiocParallel, BSgenome, caTools, DirichletMultinomial, Seqinfo, GenomicRanges, gtools, IRanges, DBI, RSQLite, rtracklayer, seqLogo, S4Vectors, TFMPvalue, XML, XVector
• Install: BiocManager::install("TFBSTools")

When to Use

• Matrix Conversion: Converting Position Frequency Matrices (PFMs) to Position Weight Matrices (PWMs) or Information Content Matrices (ICMs) using toPWM and toICM.
• Motif Scanning: Scanning nucleotide sequences or pairwise alignments for transcription factor binding sites using searchSeq or searchAln.
• Database Querying: Querying and retrieving matrix data from the JASPAR database using getMatrixSet.
• Random Profile Generation: Generating random profile matrices via permutation or Dirichlet multinomial mixture models using permuteMatrix and rPWMDmm.
• Visualization: Visualizing sequence logos for basic PWMs or Transcription Factor Flexible Models (TFFMs) using seqLogo.

When NOT to Use

• For high-throughput scanning of thousands of large genomic regions, use motifmatchr instead, because searchSeq is optimized for targeted DNAString or DNAStringSet objects rather than massive GRanges peak sets.
• For de novo motif discovery without external dependencies, use universalmotif because runMEME requires the external MEME suite to be installed on the system.

Data Requirements

• Input Format: PFMatrix objects, DNAString or DNAStringSet for sequences, and Axt objects for alignments.
• Structure: 4-row matrices representing A, C, G, T counts or probabilities.
• Normalization State: Raw position frequency counts for PFMs, which are then converted to log2 probability ratios for PWMs.