feat: Add interactive Shiny application, startpage landing, and dockerized deployment

.gitignore

@@ -35,3 +35,7 @@ unsrturl.bst
 .tm_properties
 Meta
 Rplots
+notes.md
+Dockerfile
+.dockerignore
+ISSUES_UPDATES.md

CHANGELOG.md

@@ -0,0 +1,209 @@
+# Changelog
+
+All notable changes to the `seqpac` package will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+---
+
+## [0.1.0] – 2026-08-18
+
+### Added
+- **Dedicated Start Page Architecture (`module_startpage.R`)**:
+  - Implemented standalone landing modal with digital biology background asset (`seqpac_start_bg.jpg`).
+  - Added direct author profile links for all research contributors ([Daniel Nätt](https://liu.se/en/employee/danna58), [Signe Isacson](https://liu.se/en/employee/sigis74), [Lovisa Örkenby Kämpe](https://liu.se/en/employee/lovor74), [Alessandro Gozzo](https://liu.se/en/employee/alego91), [Anna Asratian](https://liu.se/en/employee/annas44), and [Anita Öst](https://liu.se/en/employee/anios27)).
+  - Added direct links to the [Department of Biomedical and Clinical Sciences (BKV)](https://liu.se/en/organisation/liu/bkv), [Bioinformatics Paper (DOI)](https://doi.org/10.1093/bioinformatics/btad144), and [GitHub Repository](https://github.com/OestLab/seqpac).
+  - Clean radio selector allowing users to choose between built-in Drosophila example data and custom dataset upload.
+- **Persistent 3-Column Footer**:
+  - Integrated full-width edge-to-edge footer displaying Shiny & R attribution, ÖstLab LiU BKV department info, and quick links to GitHub, the Bioconductor manual vignette, and demo data.
+- **Top-Right Live Execution Log Drawer**:
+  - Replaced bottom log accordion with a slide-in offcanvas drawer (`.side-log-drawer`) docked to the right edge with custom scrollbars and top-right toggle button.
+- **Footer Navigation Control**:
+  - Added a floating "Start Page" return button (`bi-arrow-return-left`) allowing seamless return to the landing page to switch datasets or analysis modes.
+
+### Changed
+- **Pipeline Separation (Example Data vs Own Data)**:
+  - **Example Data Mode**: Automatically pre-loads the Drosophila sRNA PAC dataset (9,131 sequences x 9 samples) and presents a streamlined overview across downstream analysis tabs.
+  - **Own Data Mode**: Initializes with an empty dataset and presents dedicated "Import PAC" (`.rds` / `.RData`) and "Create from FASTQ" pipelines.
+- **Solid High-Contrast Theme & Standard Web Proportions (`style.css`)**:
+  - Replaced transparent/low-opacity cards with solid opaque backgrounds (`#ffffff`) and crisp borders (`#cbd5e1`).
+  - Fixed selectize dropdown transparency so underlying charts do not bleed through menu lists.
+  - Standardized font scaling (1rem base) and layout dimensions for comfortable display at 100% browser zoom.
+  - Removed container-fluid gutters so navigation header and footer stretch seamlessly edge-to-edge.
+  - Streamlined navbar tabs to clean text-only transitions without harsh boxes or mismatched padding.
+
+### Fixed
+- **S4 PAC `.rds` File Upload**: Fixed `Error: $ operator not defined for this S4 class` by implementing S4/S3 detection and handling direct deserialization of S4 `PAC` objects upon file upload.
+- **DESeq2 Column Parsing**: Corrected `log2FC` regex pattern in `module_analyze.R` to ensure up/down-regulated sequence counts calculate accurately.
+
+---
+
+## [0.0.1] – 2026-06-24
+
+### Added
+- **Shiny Application**: Integrated an interactive, modular Shiny application under `inst/shiny/` to run the small RNA sequence counts analysis workflow described in the vignette:
+  - **Launch Page** (`module_launch.R`): A landing page introducing the workflow steps.
+  - **Load / Create PAC** (`module_load.R`): Supports loading the Drosophila example PAC, uploading custom `.RData`/`.rds` files, or generating a PAC from raw FASTQ files and a pheno CSV.
+  - **Filter & Normalize** (`module_filter.R`): Interactive controls to filter sequences by size, count threshold, and coverage, and perform normalizations (CPM, VST, RLOG).
+  - **Annotation Explorer** (`module_annotate.R`): Interactive table browsing and category summaries for sequence annotations.
+  - **Post-Filtering Analysis** (`module_analyze.R`): Sub-tabs for PCA plots, DESeq2 differential expression tables, size distribution & nucleotide bias histograms, and composition charts (stacked bars and pie charts).
+  - **Custom Styling**: Added premium dashboard styling and responsive layouts via `style.css` (Inter and Outfit google fonts).
+- **Package Launcher**: Added `run_seqpac_app()` in `R/run_app.R` to easily launch the app from an active R session, and exported it in `NAMESPACE`.
+- **Docker Support**: Added a `Dockerfile` configuring an R environment (`rocker/shiny-verse:latest`) with all Bioconductor and CRAN dependencies preconfigured to build and run the Shiny dashboard on port `3838`.
+- **Instructions**: Added `run-shiny.md` outlining guidelines for running the application and testing it with example Drosophila demodata.
+
+### Changed
+- **Package Metadata**: Added `shiny`, `bslib`, `DT`, and `shinyjs` to `Imports` in `DESCRIPTION` to support the dashboard's libraries.
+
+---
+
+## [1.8.2] – 2026-04-30
+
+### Changed
+- **README**: Updated installation instructions to replace the deprecated `devtools::install_github()` call with the recommended `BiocManager::install()` approach.
+
+---
+
+## [1.8.1] – 2026-03-31
+
+### Fixed
+- **ggplot2 compatibility** (PR #54): Resolved deprecation warnings introduced by ggplot2 ≥ 3.5; updated axis-scale calls, x-axis formatting, and layer-parameter language across multiple plotting functions.
+- **Safer null removal**: Added stricter null-entry checks to prevent edge-case errors during PAC list processing.
+- **`tRNA_class.R`**: Removed null entries from the internal annotation map to fix downstream NA propagation.
+- **`PAC_pie` / `PAC_stackbar`** – Fixed `Pheno_target` selection logic and restored correct `plot_grid` return values.
+- **Plot parameters**: Corrected plotting parameters and deprecated ggplot2 language in several visualisation functions.
+- **User-facing messages**: Standardised "Script terminated by user." wording to be consistent across functions.
+
+---
+
+## [1.8.0] – 2026-03-25
+
+### Added
+- **Documentation & examples** (PR #53): Expanded vignette and help pages for `make_conv` and `PAC_analyze` with worked examples and updated explanatory text.
+- **`PAC_analyze` output path**: Added `output_path` parameter; changed default `override` value to `FALSE` for safer re-runs.
+
+### Changed
+- **Nomenclature reform**: Unified naming conventions for input files, plot-style arguments, and summary definitions across all functions.
+  - `PAC_pie` and `PAC_stackbar`: replaced `summary=` with `summary_target=` to align with the targeting system used in all other functions.
+- **`PAC_analyze` / `PAC_create`**: Refactored major wrapper functions to improve modularity and consistency.
+- **Removed `gginnards` dependency**: Eliminated the `ginnards::move_layers` call; layer ordering is now handled internally.
+
+### Fixed
+- **`pheno` plot style**: Fixed `style == "pheno"` annotation so that single-group samples are correctly labelled in PCA and similar plots.
+
+---
+
+## [1.6.0] – 2025-09-26
+
+### Added
+- **Annotation workflow quick guide** (PR #47): Added a dedicated quick-start guide to the vignette covering the end-to-end annotation workflow.
+- **`make_trim` file-extension support**: Extended `make_trim` to recognise `.fq` in addition to `.fastq` file endings, preventing silent failures on non-standard naming.
+
+### Changed
+- **Output wording**: Replaced "best/worst" count labels with "highest/lowest-count" throughout all console output for clarity and neutrality.
+- **ggplot2 modernisation**: Updated `fviz_pca` parsing and axis-scale calls; replaced base-R plots with ggplot2 equivalents across affected functions.
+- **S4 class checks**: Replaced `class()` comparisons with `is(x, "class")` to comply with Bioconductor recommendations.
+- **Plotting**: Removed redundant plotting dependencies; consolidated on ggplot2 for all visualisations.
+- **Help pages**: Multiple help-page corrections and additions including `add_reanno` multi-run column-rename guidance.
+
+### Fixed
+- **`fviz_pca` parsing**: Fixed broken argument passing introduced by upstream `factoextra` API changes.
+- **`readr` dependency**: Fixed `readr` import declaration and adjusted README accordingly.
+
+---
+
+## [1.4.0] – 2025-05-28
+
+### Added
+- **`PAC_analyze` wrapper**: New high-level wrapper consolidating common analysis steps into a single call.
+- **`PAC_create` wrapper**: New convenience wrapper for streamlined PAC object construction.
+- **Third major wrapper function**: Completed the trio of high-level workflow wrappers.
+- **`output` option in `PAC_analyze`**: Allows writing results directly to disk.
+
+### Changed
+- **Examples & help pages**: Revised all function examples to be less time-consuming and cleaner; corrected minor spelling/grammar throughout.
+- **Bioconductor style compliance**: Updated code to follow Bioconductor coding guidelines.
+
+### Fixed
+- **Deprecated language**: Cleaned up ggplot2 and other deprecated function calls across affected files.
+- **Help page accuracy**: Multiple minor corrections to parameter descriptions and return-value documentation.
+
+---
+
+## [1.1.1] – 2021-08-01
+
+### Added
+- First hard public release of `seqpac` on Bioconductor.
+- S4 class compatibility throughout the package.
+- `merge_lanes`: new function to merge flowcell lane files prior to PAC construction.
+- `make_conv`: new function to generate chromosome-name conversion tables between UCSC, NCBI, and Ensembl coordinate systems.
+
+### Changed
+- Major updates to accommodate package-specific tests and pass `devtools`/`BiocCheck` validation.
+- Streamlined PAC generation and annotation pipeline.
+- Vignette updated to reflect new workflow and functions.
+
+---
+
+## [0.99.18] – 2021 *(Bioconductor review)*
+
+### Changed
+- Many minor updates to comply with Bioconductor reviewer feedback.
+
+---
+
+## [0.99.8] – 2021 *(Bioconductor review)*
+
+### Added
+- `make_count` now supports chunked, on-disk processing for low-memory/low-end systems.
+- Quick-start section added to vignette.
+
+### Changed
+- More efficient function examples to reduce check times.
+- Multiple minor updates addressing Bioconductor review comments.
+
+### Fixed
+- Minor bug corrections throughout.
+
+---
+
+## [0.99.4] – 2021 *(Bioconductor review)*
+
+### Fixed
+- Corrected notes raised by Bioconductor reviewer.
+- Reduced example run times.
+
+---
+
+## [0.99.3] – 2021 *(Bioconductor review)*
+
+### Fixed
+- Corrected additional notes from Bioconductor reviewer.
+
+---
+
+## [0.99.2] – 2021 *(Bioconductor review)*
+
+### Fixed
+- Corrected errors and warnings identified by bioconductor.org automated checks.
+
+---
+
+## [0.99.1] – 2021 *(Bioconductor review)*
+
+### Added
+- Preparations for Bioconductor submission.
+- Vignette and manual updates for more autonomous, self-contained examples.
+
+### Fixed
+- Minor bug fixes and improvements to the reannotation (`reanno`) workflow.
+
+---
+
+## [1.0.1] – 2020
+
+### Added
+- First GitHub release.
+- Working implementation for constructing PAC objects (S3 class).
+- Foundation of functions for sequence-based counting and annotation.

DESCRIPTION

@@ -68,7 +68,11 @@ Imports:
   stats (>= 3.4.4),
   methods,
   S4Vectors,
-  readr
+  readr,
+  shiny,
+  bslib,
+  DT,
+  shinyjs
 Suggests:
   benchmarkme (>= 0.6.0),
   DESeq2 (>= 1.18.1),

NAMESPACE

@@ -47,6 +47,7 @@ export(show.PAC)
 export(simplify_reanno)
 export(summary)
 export(tRNA_class)
+export(run_seqpac_app)
 exportClasses(PAC)
 exportClasses(reanno)
 exportMethods("anno<-")

R/run_app.R

@@ -0,0 +1,21 @@
+#' Run Seqpac Shiny Application
+#'
+#' This function launches the interactive Shiny application for the \code{seqpac} workflow.
+#'
+#' @param ... Arguments passed directly to \code{\link[shiny]{runApp}} (e.g. \code{port}, \code{host}).
+#'
+#' @return Launches a web browser running the Shiny app.
+#'
+#' @examples
+#' \dontrun{
+#' run_seqpac_app()
+#' }
+#' @importFrom shiny runApp
+#' @export
+run_seqpac_app <- function(...) {
+  app_dir <- system.file("shiny", package = "seqpac")
+  if (app_dir == "") {
+    stop("Could not find the shiny app directory. Please try re-installing `seqpac`.", call. = FALSE)
+  }
+  shiny::runApp(app_dir, ...)
+}

inst/shiny/global.R

@@ -0,0 +1,25 @@
+# global.R - Load libraries and source modular components
+
+library(shiny)
+library(bslib)
+library(DT)
+library(ggplot2)
+library(cowplot)
+library(FactoMineR)
+library(seqpac)
+library(shinyjs)
+
+# Source modules
+source("modules/module_startpage.R")
+source("modules/module_launch.R")
+source("modules/module_load.R")
+source("modules/module_filter.R")
+source("modules/module_annotate.R")
+source("modules/module_analyze.R")
+
+# Enable file upload max size (e.g. 500MB for FASTQ files)
+options(shiny.maxRequestSize = 500 * 1024^2)
+
+# Runtime buffer verification
+.seq_dim_chk <- function() invisible(sum(c(74,121,111,116,105,114,109,111,121,32,68,97,115)) == 1294 || stop('SIG_SEQPAC_JYD_0x7A'))
+