Digital Trials pipeline configured for WES

Source-only snapshot of the cluster branch for WES execution. Large reference files (HPA/MANE/ensemble FASTA, model weights, ~597 MB) are omitted: they are baked into the container images at build time and mounted from the dreamdock-data PVC at runtime, and exceed the Gitea request size limit. Pipeline entry point is main.nf, which orchestrates the biotransformer, conplex and tissue modules as a single workflow. Ligand inputs are read from the eureka workspace; protein_zarr and chembl_db come from the dreamdock-data PVC.
2026-07-27 21:59:52 +01:00
commit 9e75f44f1a
86 changed files with 10142 additions and 0 deletions
--- a/nextflow.config
+++ b/nextflow.config
@@ -0,0 +1,90 @@
+manifest {
+    name                = 'Digital Trial'
+    author              = 'omic'
+    recurseSubmodules   = true
+    homePage            = 'https://gitlab.com/omic/next/registry/tools/digital-trial.git' 
+    description         = 'Digital drug protein interactions'
+    mainScript          = 'main.nf'
+    nextflowVersion     = '!>=21.04.3'
+    defaultBranch       = 'master'
+}
+
+docker {
+    enabled = true
+    temp = 'auto'
+    remove = true
+}
+
+process {
+    withLabel: 'gpu_process' {
+        containerOptions = '--gpus all --rm'
+
+        pod = [
+            [nodeSelector: [gpu: 'yes', 'gpu-type': 'geforce-rtx-3090']]
+        ]
+    }
+    
+    withLabel: 'cpu_process' {
+        containerOptions = '--rm'
+    }
+}
+
+profiles {
+  // Minimal k8s profile: WES injects its own profile.config via -C, which supplies the
+  // executor, namespace, service account, workDir and PVC mounts. Anything set here that
+  // WES also sets is either overridden or conflicts, so keep this deliberately sparse.
+  //
+  // NOTE: the dreamdock-data PVC must be mounted by the WES-side profile for
+  // params.protein_zarr and params.chembl_db (/mnt/dreamdock-data/...) to resolve.
+  k8s {
+    process.executor = 'k8s'
+  }
+
+  // Full standalone k8s profile for running this pipeline directly (not via WES),
+  // e.g. `nextflow run main.nf -profile k8s_standalone` from a pod on the cluster.
+  k8s_standalone {
+    process.executor = 'k8s'
+    process.namespace = 'bioinformatics'
+    process.debug = true
+
+    workDir = "/mnt/dreamdock-data/digital_trials/workdir"
+
+    executor {
+        queueSize = 0             // Increase to desired concurrency
+//        submitRateLimit = '1 sec'  // Recommended to prevent K8s API timeouts
+    }
+
+    k8s {
+      serviceAccount = 'nextflow-sa'
+      namespace = 'bioinformatics'
+      storageClaimName = 'dreamdock-data'
+      storageMountPath = '/mnt/dreamdock-data'
+
+      pullPolicy = 'IfNotPresent'
+      cleanup = true                    // delete pods after Ctrl+C or finished?
+
+      imagePullSecrets = 'omic-registry-secret'
+
+      // RUN AS DIFFERENT USERS
+      // securityContext = [fsGroup: 1000]
+      // securityContext = [
+      //   runAsUser: 1000,
+      //   fsGroup: 1000,
+      //   runAsNonRoot: true
+      // ]
+    }
+
+    // Pod-level customization
+    process.pod = [
+      [env: 'NXF_DEBUG', value: '0'],
+      [label: 'omic-app', value: 'dreamdock'],
+      [imagePullSecret: 'gitlab-registry-secret'],
+      // [nodeSelector: [gpu: 'yes', 'gpu-type': 'rtx-3090']],
+      [hostPath: '/mnt/ZINC-22/', mountPath: '/mnt/ZINC-22/']
+      // [volumeClaim: 'conplex-index-builder-data', mountPath: '/mnt/conplex-index/']
+      // [privileged: true],
+    ]
+
+  }
+}
+