Merge pull request #185819 from apraga/gatk-4.2.6.1

maintainers/maintainer-list.nix

@@ -1027,6 +1027,12 @@
       fingerprint = "BF8B F725 DA30 E53E 7F11  4ED8 AAA5 0652 F047 9205";
     }];
   };
+  apraga = {
+    email = "alexis.praga@proton.me";
+    github = "apraga";
+    githubId = 914687;
+    name = "Alexis Praga";
+  };
   ar1a = {
     email = "aria@ar1as.space";
     github = "ar1a";

pkgs/applications/science/biology/gatk/default.nix

@@ -0,0 +1,38 @@
+{ lib, stdenv, fetchzip }:
+
+stdenv.mkDerivation rec {
+  pname = "gatk";
+  version = "4.2.6.1";
+  src = fetchzip {
+    url = "https://github.com/broadinstitute/gatk/releases/download/${version}/gatk-${version}.zip";
+    sha256 = "0hjlsl7fxf3ankyjidqhwxc70gjh6z4lnjzw6b5fldzb0qvgfvy8";
+  };
+
+  installPhase = ''
+    mkdir -p $out/bin
+    install -m755 -D $src/gatk $out/bin/
+    install -m755 -D $src/gatk-package-${version}-local.jar $out/bin/
+  '';
+
+  meta = with lib; {
+    homepage = "https://gatk.broadinstitute.org/hc/en-us";
+    description = "A wide variety of tools with a primary focus on variant discovery and genotyping." ;
+    license = licenses.asl20;
+    sourceProvenance = with lib.sourceTypes; [ binaryBytecode ] ;
+    maintainers = with maintainers; [ apraga ];
+    longDescription = ''
+      The GATK is the industry standard for identifying SNPs and indels in germline
+      DNA and RNAseq data. Its scope is now expanding to include somatic short variant
+      calling, and to tackle copy number (CNV) and structural variation (SV). In
+      addition to the variant callers themselves, the GATK also includes many
+      utilities to perform related tasks such as processing and quality control of
+      high-throughput sequencing data, and bundles the popular Picard toolkit.
+
+      These tools were primarily designed to process exomes and whole genomes
+      generated with Illumina sequencing technology, but they can be adapted to handle
+      a variety of other technologies and experimental designs. And although it was
+      originally developed for human genetics, the GATK has since evolved to handle
+      genome data from any organism, with any level of ploidy.
+    '';
+  };
+}

pkgs/top-level/all-packages.nix

@@ -6629,6 +6629,8 @@ with pkgs;
 
   gawkInteractive = gawk.override { interactive = true; };
 
+  gatk = callPackage ../applications/science/biology/gatk {} ;
+
   gbdfed = callPackage ../tools/misc/gbdfed {
     gtk = gtk2-x11;
   };