From 7269ae4e7b4d4aa2b8e1631a216a5531eb7165b6 Mon Sep 17 00:00:00 2001
From: Theodoro Gasperin Terra Camargo
 <98555209+tgaspe@users.noreply.github.com>
Date: Mon, 2 Sep 2024 14:42:44 +0200
Subject: [PATCH] Bedtools links (#137)

* Initial Commit

* Tests

* Adding help file

* Adding more description

* Update test.sh

* Update help.txt

* Update CHANGELOG.md
---
 CHANGELOG.md                                |  1 +
 src/bedtools/bedtools_links/config.vsh.yaml | 91 +++++++++++++++++++
 src/bedtools/bedtools_links/help.txt        | 25 ++++++
 src/bedtools/bedtools_links/script.sh       | 14 +++
 src/bedtools/bedtools_links/test.sh         | 98 +++++++++++++++++++++
 5 files changed, 229 insertions(+)
 create mode 100644 src/bedtools/bedtools_links/config.vsh.yaml
 create mode 100644 src/bedtools/bedtools_links/help.txt
 create mode 100644 src/bedtools/bedtools_links/script.sh
 create mode 100644 src/bedtools/bedtools_links/test.sh

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 8c1af805..6dda7ab4 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -32,6 +32,7 @@
   - `bedtools/bedtools_merge`: Merges overlapping BED/GFF/VCF entries into a single interval (PR #118).
   - `bedtools/bedtools_bamtofastq`: Convert BAM alignments to FASTQ files (PR #101).
   - `bedtools/bedtools_bedtobam`: Converts genomic feature records (bed/gff/vcf) to BAM format (PR #111).
+  - `bedtools/bedtools_links`: Creates an HTML file with links to an instance of the UCSC Genome Browser for all features / intervals in a (bed/gff/vcf) file (PR #137).
  
 * `qualimap/qualimap_rnaseq`: RNA-seq QC analysis using qualimap (PR #74). 
 
diff --git a/src/bedtools/bedtools_links/config.vsh.yaml b/src/bedtools/bedtools_links/config.vsh.yaml
new file mode 100644
index 00000000..b4e43cd3
--- /dev/null
+++ b/src/bedtools/bedtools_links/config.vsh.yaml
@@ -0,0 +1,91 @@
+name: bedtools_links
+namespace: bedtools
+description: | 
+  Creates an HTML file with links to an instance of the UCSC Genome Browser for all features / intervals in a file. 
+  This is useful for cases when one wants to manually inspect through a large set of annotations or features.
+keywords: [Links, BED, GFF, VCF]
+links:
+  documentation: https://bedtools.readthedocs.io/en/latest/content/tools/links.html
+  repository: https://github.com/arq5x/bedtools2
+  homepage: https://bedtools.readthedocs.io/en/latest/#
+  issue_tracker: https://github.com/arq5x/bedtools2/issues
+references:
+  doi: 10.1093/bioinformatics/btq033
+license: MIT
+requirements:
+  commands: [bedtools]
+authors:
+  - __merge__: /src/_authors/theodoro_gasperin.yaml
+    roles: [ author, maintainer ]
+
+argument_groups:
+  - name: Inputs
+    arguments:
+      - name: --input
+        alternatives: -i
+        type: file
+        description: Input file (bed/gff/vcf).
+        required: true
+    
+  - name: Outputs
+    arguments:
+      - name: --output
+        alternatives: -o
+        type: file
+        direction: output
+        description: Output HTML file to be written.
+
+  - name: Options
+    description: |
+      By default, the links created will point to human (hg18) UCSC browser.
+      If you have a local mirror, you can override this behavior by supplying
+      the -base, -org, and -db options.
+
+      For example, if the URL of your local mirror for mouse MM9 is called: 
+      http://mymirror.myuniversity.edu, then you would use the following:
+      --base_url http://mymirror.myuniversity.edu
+      --organism mouse
+      --database mm9
+    arguments:
+      - name: --base_url
+        alternatives: -base
+        type: string
+        description: | 
+          The “basename” for the UCSC browser.
+        default: http://genome.ucsc.edu
+      
+      - name: --organism
+        alternatives: -org
+        type: string
+        description: | 
+          The organism (e.g. mouse, human). 
+        default: human
+
+      - name: --database
+        alternatives: -db
+        type: string
+        description: | 
+          The genome build. 
+        default: hg18
+      
+resources:
+  - type: bash_script
+    path: script.sh
+
+test_resources:
+  - type: bash_script
+    path: test.sh
+
+engines:
+  - type: docker
+    image: debian:stable-slim
+    setup:
+      - type: apt
+        packages: [bedtools, procps]
+      - type: docker
+        run: |
+          echo "bedtools: \"$(bedtools --version | sed -n 's/^bedtools //p')\"" > /var/software_versions.txt
+
+runners:
+  - type: executable
+  - type: nextflow
diff --git a/src/bedtools/bedtools_links/help.txt b/src/bedtools/bedtools_links/help.txt
new file mode 100644
index 00000000..d848d989
--- /dev/null
+++ b/src/bedtools/bedtools_links/help.txt
@@ -0,0 +1,25 @@
+```
+bedtools links -h
+```
+
+Tool:    bedtools links (aka linksBed)
+Version: v2.30.0
+Summary: Creates HTML links to an UCSC Genome Browser from a feature file.
+
+Usage:   bedtools links [OPTIONS] -i <bed/gff/vcf> > out.html
+
+Options: 
+	-base	The browser basename.  Default: http://genome.ucsc.edu 
+	-org	The organism. Default: human
+	-db	The build.  Default: hg18
+
+Example: 
+	By default, the links created will point to human (hg18) UCSC browser.
+	If you have a local mirror, you can override this behavior by supplying
+	the -base, -org, and -db options.
+
+	For example, if the URL of your local mirror for mouse MM9 is called: 
+	http://mymirror.myuniversity.edu, then you would use the following:
+	-base http://mymirror.myuniversity.edu
+	-org mouse
+	-db mm9
diff --git a/src/bedtools/bedtools_links/script.sh b/src/bedtools/bedtools_links/script.sh
new file mode 100644
index 00000000..b8ee9a56
--- /dev/null
+++ b/src/bedtools/bedtools_links/script.sh
@@ -0,0 +1,14 @@
+#!/bin/bash
+
+## VIASH START
+## VIASH END
+
+set -eo pipefail
+
+# Execute bedtools links
+bedtools links \
+    ${par_base_url:+-base "$par_base_url"} \
+    ${par_organism:+-org "$par_organism"} \
+    ${par_database:+-db "$par_database"} \
+    -i "$par_input" \
+    > "$par_output"
diff --git a/src/bedtools/bedtools_links/test.sh b/src/bedtools/bedtools_links/test.sh
new file mode 100644
index 00000000..d79cbd6c
--- /dev/null
+++ b/src/bedtools/bedtools_links/test.sh
@@ -0,0 +1,98 @@
+#!/bin/bash
+
+# exit on error
+set -eo pipefail
+
+#############################################
+# helper functions
+assert_file_exists() {
+  [ -f "$1" ] || { echo "File '$1' does not exist" && exit 1; }
+}
+assert_file_not_empty() {
+  [ -s "$1" ] || { echo "File '$1' is empty but shouldn't be" && exit 1; }
+}
+assert_file_contains() {
+  grep -q "$2" "$1" || { echo "File '$1' does not contain '$2'" && exit 1; }
+}
+assert_identical_content() {
+  diff -a "$2" "$1" \
+    || (echo "Files are not identical!" && exit 1)
+}
+#############################################
+
+# Create directories for tests
+echo "Creating Test Data..."
+TMPDIR=$(mktemp -d "$meta_temp_dir/XXXXXX")
+function clean_up {
+  [[ -d "$TMPDIR" ]] && rm -r "$TMPDIR"
+}
+trap clean_up EXIT
+
+# Create test data
+cat <<EOF > "$TMPDIR/genes.bed"
+chr21	9928613	10012791	uc002yip.1	0	-
+chr21	9928613	10012791	uc002yiq.1	0	-
+chr21	9928613	10012791	uc002yir.1	0	-
+chr21	9928613	10012791	uc010gkv.1	0	-
+chr21	9928613	10061300	uc002yis.1	0	-
+chr21	10042683	10120796	uc002yit.1	0	-
+chr21	10042683	10120808	uc002yiu.1	0	-
+chr21	10079666	10120808	uc002yiv.1	0	-
+chr21	10080031	10081687	uc002yiw.1	0	-
+chr21	10081660	10120796	uc002yix.2	0	-
+EOF
+
+# Test 1: Default Use
+mkdir "$TMPDIR/test1" && pushd "$TMPDIR/test1" > /dev/null
+
+echo "> Run bedtools_links on BED file"
+"$meta_executable" \
+  --input "../genes.bed" \
+  --output "genes.html"
+
+# checks
+assert_file_exists "genes.html"
+assert_file_not_empty "genes.html"
+assert_file_contains "genes.html" "uc002yip.1"
+echo "- test1 succeeded -"
+
+popd > /dev/null
+
+# Test 2: Base URL
+mkdir "$TMPDIR/test2" && pushd "$TMPDIR/test2" > /dev/null
+
+echo "> Run bedtools_links with base option"
+"$meta_executable" \
+  --input "../genes.bed" \
+  --output "genes.html" \
+  --base_url "http://genome.ucsc.edu"
+
+# checks
+assert_file_exists "genes.html"
+assert_file_not_empty "genes.html"
+assert_file_contains "genes.html" "uc002yip.1"
+echo "- test2 succeeded -"
+
+popd > /dev/null
+
+# Test 3: Organism and Genome Database Build
+mkdir "$TMPDIR/test3" && pushd "$TMPDIR/test3" > /dev/null
+
+echo "> Run bedtools_links with organism option and genome database build"
+"$meta_executable" \
+  --input "../genes.bed" \
+  --output "genes.html" \
+  --base_url "http://genome.ucsc.edu" \
+  --organism "mouse" \
+  --database "mm9"
+
+# checks
+assert_file_exists "genes.html"
+assert_file_not_empty "genes.html"
+assert_file_contains "genes.html" "uc002yip.1"
+echo "- test3 succeeded -"
+
+popd > /dev/null
+
+echo "---- All tests succeeded! ----"
+exit 0