# Bindingsitefinder

> BindingSiteFinder

- Skill: `biomate-ai/bindingsitefinder` (Agent Skill)
- Install (CLI): `npx skillmds@latest add biomate-ai/bindingsitefinder`
- Raw SKILL.md: https://api.skillmd.com/api/skills/biomate-ai/bindingsitefinder/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: biomate-ai (https://skillmd.com/u/biomate-ai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/biomate-ai/bindingsitefinder

---


# BindingSiteFinder

## Workflows

### Standard Workflow

```r
library(BindingSiteFinder)
library(txdbmaker)
library(rtracklayer)

# 1. Import crosslink sites
csFile <- system.file("extdata", "PureCLIP_crosslink_sites_examples.bed", package="BindingSiteFinder")
cs <- rtracklayer::import(con = csFile, format = "BED")

# 2. Create metadata table and construct BSFDataSet
files <- system.file("extdata", package="BindingSiteFinder")
clipFilesP <- list.files(files, pattern = "plus.bw$", full.names = TRUE)
clipFilesM <- list.files(files, pattern = "minus.bw$", full.names = TRUE)
meta <- data.frame(
  id = 1:4,
  condition = factor(rep("WT", 4)),
  clPlus = clipFilesP,
  clMinus = clipFilesM
)
bds <- BSFDataSetFromBigWig(ranges = cs, meta = meta, silent = TRUE)

# 3. Build gene and transcript region annotations
annoDb <- txdbmaker::makeTxDbFromGFF(file = "gencode_v37_annotation.gff3", format = "gff3")
gns <- genes(annoDb)
regions <- GRangesList(
  CDS = cds(annoDb),
  Intron = unlist(intronsByTranscript(annoDb)),
  UTR3 = unlist(threeUTRsByTranscript(annoDb)),
  UTR5 = unlist(fiveUTRsByTranscript(annoDb))
)

# 4. Run BSFind wrapper
bdsOut <- BSFind(
  object = bds,
  anno.genes = gns,
  anno.transcriptRegionList = regions,
  est.subsetChromosome = "chr22",
  veryQuiet = TRUE
)

# 5. Export results
exportToBED(bdsOut, con = "./myBindingSites.bed")
```
**Input/Output Note:** Inputs a `BSFDataSet` and genomic annotations; outputs a refined `BSFDataSet` with defined binding sites and exports them to a BED file.

### Differential Binding Analysis

```r
library(BindingSiteFinder)
library(txdbmaker)

# 1. Load Gene Annotations and Transcript Regions
annoDb <- txdbmaker::makeTxDbFromGFF(file = "gencode_v37_annotation.gff3", format = "gff3")
gns <- genes(annoDb)
regions <- GRangesList(
  CDS = cds(annoDb),
  Intron = unlist(intronsByTranscript(annoDb)),
  UTR3 = unlist(threeUTRsByTranscript(annoDb)),
  UTR5 = unlist(fiveUTRsByTranscript(annoDb))
)

# 2. Define binding sites for a condition
bdsOut <- BSFind(
  object = bds, 
  anno.genes = gns, 
  anno.transcriptRegionList = regions, 
  est.subsetChromosome = "chr22"
)
```
**Input/Output Note:** Inputs a `BSFDataSet` and genomic annotations; outputs a `BSFDataSet` with defined binding sites prepared for downstream differential testing.

## When to Use
- To define precise, equally sized RBP binding sites from genome-wide iCLIP coverage and peak calling results (e.g., PureCLIP crosslink sites).
- To filter, merge, and check reproducibility of crosslink sites across replicates using `BSFind()`.
- To annotate binding sites with genomic features such as genes and transcript regions (CDS, Introns, UTRs) using `genes()`, `cds()`, and `intronsByTranscript()`.

## When NOT to Use
- For analyzing RNA-seq differential expression or splicing directly without iCLIP/CLIP-seq data (use `DESeq2` or `dexseq` instead).
- For peak calling directly from BAM files (use peak callers like `PureCLIP` first, then import results into `BindingSiteFinder`).

## Data Requirements
- Crosslink sites as a `GRanges` object (typically single-nucleotide wide, e.g., imported from PureCLIP BED files).
- Replicate-specific coverage data in strand-specific BigWig files (`clPlus` and `clMinus`).
- Gene annotations as a `GRanges` object and transcript regions as a `GRangesList` (e.g., CDS, Introns, 3' UTR, 5' UTR).

## Key Parameters
- **object**: A `BSFDataSet` containing crosslink ranges and metadata.
- **anno.genes**: A `GRanges` object containing gene annotations.
- **anno.transcriptRegionList**: A `GRangesList` containing transcript region annotations.
- **est.subsetChromosome** ("chr22"): Character vector specifying which chromosome to use for estimating binding site width.
- **veryQuiet** (FALSE): Logical indicating whether to suppress progress messages.
- **est.maxBsWidth** (29): Numeric specifying the maximum binding site width to test during estimation.

## Best Practices
- Pre-filter crosslink sites with low scores (e.g., removing the lowest 1% of PureCLIP scores) using `pureClipGlobalFilterPlot()` to assess the cutoff.
- Use `estimateBsWidthPlot()` to evaluate the optimal binding site width based on the signal-to-flank ratio.
- Enforce reproducibility across replicates by setting the `nReps` parameter (typically N-1 replicates) and verifying with `reproducibilityScatterPlot()`.

## Common Pitfalls
- *Pitfall*: Crosslink sites overlapping multiple genes causing inflated site counts. *Fix*: Set `overlaps = "keepSingle"` in `BSFind()` to reduce overlapping loci to a single instance.
- *Pitfall*: Low reproducibility across replicates due to library size differences. *Fix*: Use `reproducibilityFilterPlot()` to inspect and adjust the replicate-specific crosslink threshold.

## Alternatives
- `csaw` for window-based differential binding analysis of ChIP-seq/ATAC-seq.
- `macsr` for peak calling on enrichment data.
- `CLIPreg` for integrating CLIP-seq with RNA-seq.

## Citations
- Busch A, Brüggemann M, Ebersberger S, Zarnack K (2020). "iCLIP data analysis: A complete pipeline from sequencing reads to RBP binding sites." *Methods*, 178, 49-62. doi:10.1016/j.ymeth.2019.11.008.

## References
- Homepage: https://bioconductor.org/packages/bindingsitefinder
- Vignette: https://bioconductor.org/packages/release/bioc/vignettes/bindingsitefinder/inst/doc/Definition_of_binding_sites_from_iCLIP_signal.html