Add FiberBrowser page to Dry Lab section

src/README.md

@@ -21,6 +21,7 @@ The method is described in:
 
 **Dry Lab**
 - [DAF-QC Pipeline](drylab/daf-qc.md) -- Snakemake pipeline for quality control and initial processing of DAF-seq data
+- [FiberBrowser](drylab/fiberbrowser.md) -- Interactive single-molecule genome browser for DAF-seq and Fiber-seq data
 
 **Reference**
 - [Glossary](glossary.md) -- Definitions of key DAF-seq terms and concepts

src/SUMMARY.md

@@ -17,6 +17,7 @@
 # Dry Lab
 
 - [DAF-QC Pipeline](drylab/daf-qc.md)
+- [FiberBrowser](drylab/fiberbrowser.md)
 
 ---
 

src/drylab/fiberbrowser.md

@@ -0,0 +1,56 @@
+# FiberBrowser
+
+FiberBrowser is an interactive single-molecule genome browser for DAF-seq
+and Fiber-seq data, and the visualization component of the
+[FiberHMM](https://github.com/mtcicero26/FiberHMM_v2) pipeline. It is built
+for labs that want to explore per-read footprint patterns interactively at
+gene-scale loci -- the regime where single-molecule chromatin assays diverge
+most from bulk tracks -- and that care about analyzing the full read set at
+a locus rather than a downsampled view.
+
+![FiberBrowser overview at the eve locus](img/fiberbrowser-overview.png)
+
+*2-4 hr Fiber-seq at the* eve *locus in Drosophila embryos. 1,259 reads span
+the region; the browser draws every footprint on every read, and every
+downstream analysis is computed on the full read set.*
+
+## Features
+
+- **Per-read heatmaps** of m6A / DAF accessibility, drawn as a single canvas
+  layer so loci with thousands of reads stay responsive.
+- **Single-molecule clustering** on CRE accessibility patterns, with
+  per-cluster metaprofiles, merge / reorder / hide controls, and a sortable
+  read table.
+- **Direct comparison across perturbations or timepoints** -- load multiple
+  datasets at once and joint-cluster them so the same single-molecule state
+  definitions apply across conditions, making perturbation effects readable
+  at single-read resolution.
+- **Customizable footprint labeling** -- recolor footprints by user-defined
+  rules (Pol II, NFR, TF, nucleosome) using size and overlap criteria.
+- **Transcription-state quantification** -- filter and count reads matching
+  paused, PIC, elongating, or hyperbound Pol II configurations at a
+  selected gene.
+- **Footprint inspection and motif enrichment** -- pull motif hits and
+  FASTA for any single footprint and compare to references (ChIP-nexus,
+  motif databases).
+- **DAF-seq primer design** -- design and BLAST-check primers for targeted
+  DAF-seq amplicons directly from the browser, with IDT / Twist CSV export.
+  See also the [Primer Design](../protocol/primer-design.md) guidelines.
+- **Headless / API mode and Jupyter control** -- drive the browser
+  programmatically via its FastAPI endpoints or a Python client; load
+  regions, pull cluster assignments, fetch co-accessibility matrices, and
+  render figures directly from a notebook.
+- **Local LLM copilot** (optional) -- natural-language commands that drive
+  the browser via a model running on your machine (MLX on Mac, Ollama
+  elsewhere); doubles as an interactive tutorial for discovering features.
+- ...and more.
+
+FiberBrowser sits downstream of the [DAF-QC Pipeline](daf-qc.md): once you
+have QC'd reads and called footprints, FiberBrowser is where you explore
+them.
+
+## Early access
+
+FiberBrowser is in pre-release while we finalize the public repository. If
+you'd like to try it on your data, reach out at tommywtullius [at]
+gmail.com.