Add sentieon star

.github/actions/nf-test/action.yml

@@ -54,10 +54,20 @@ runs:
         conda-solver: libmamba
         conda-remove-defaults: true
 
+    # Set up secrets
+    - name: Set up Nextflow secrets
+      if: env.SENTIEON_ENCRYPTION_KEY != '' && env.SENTIEON_LICENSE_MESSAGE != ''
+      shell: bash
+      run: |
+        python -m pip install cryptography
+        nextflow secrets set SENTIEON_AUTH_DATA $(python3 .github/actions/nf-test/license_message.py encrypt --key "$SENTIEON_ENCRYPTION_KEY" --message "$SENTIEON_LICENSE_MESSAGE")
+
     - name: Run nf-test
       shell: bash
       env:
         NFT_WORKDIR: ${{ env.NFT_WORKDIR }}
+        SENTIEON_LICSRVR_IP: ${{ env.SENTIEON_LICSRVR_IP }}
+        SENTIEON_AUTH_MECH: "GitHub Actions - token"
       run: |
         nf-test test \
           --profile=+${{ inputs.profile }} \

.github/actions/nf-test/license_message.py

@@ -0,0 +1,113 @@
+#!/usr/bin/env python3
+
+#########################################
+# Author: [DonFreed](https://github.com/DonFreed)
+# File: license_message.py
+# Source: https://github.com/DonFreed/docker-actions-test/blob/main/.github/scripts/license_message.py
+# Source+commit: https://github.com/DonFreed/docker-actions-test/blob/aa1051a9f53b3a1e801953748d062cad74dca9a9/.github/scripts/license_message.py
+# Download Date: 2023-07-04, commit: aa1051a
+# This source code is licensed under the BSD 2-Clause license
+#########################################
+
+"""
+Functions for generating and sending license messages
+"""
+
+# Modified from - https://stackoverflow.com/a/59835994
+
+import argparse
+import base64
+import calendar
+import re
+import secrets
+import sys
+
+from cryptography.hazmat.primitives.ciphers.aead import AESGCM
+from datetime import datetime as dt
+
+MESSAGE_TIMEOUT = 60 * 60 * 24  # Messages are valid for 1 day
+NONCE_BYTES = 12
+
+
+class DecryptionTimeout(Exception):
+    # Decrypting a message that is too old
+    pass
+
+
+def generate_key():
+    key = secrets.token_bytes(32)
+    return key
+
+
+def handle_generate_key(args):
+    key = generate_key()
+    key_b64 = base64.b64encode(key)
+    print(key_b64.decode("utf-8"), file=args.outfile)
+
+
+def encrypt_message(key, message):
+    nonce = secrets.token_bytes(NONCE_BYTES)
+    timestamp = calendar.timegm(dt.now().utctimetuple())
+    data = timestamp.to_bytes(10, byteorder="big") + b"__" + message
+    ciphertext = nonce + AESGCM(key).encrypt(nonce, data, b"")
+    return ciphertext
+
+
+def handle_encrypt_message(args):
+    key = base64.b64decode(args.key.encode("utf-8"))
+    message = args.message.encode("utf-8")
+    ciphertext = encrypt_message(key, message)
+    ciphertext_b64 = base64.b64encode(ciphertext)
+    print(ciphertext_b64.decode("utf-8"), file=args.outfile)
+
+
+def decrypt_message(key, ciphertext, timeout=MESSAGE_TIMEOUT):
+    nonce, ciphertext = ciphertext[:NONCE_BYTES], ciphertext[NONCE_BYTES:]
+    message = AESGCM(key).decrypt(nonce, ciphertext, b"")
+
+    msg_timestamp, message = re.split(b"__", message, maxsplit=1)
+    msg_timestamp = int.from_bytes(msg_timestamp, byteorder="big")
+    timestamp = calendar.timegm(dt.now().utctimetuple())
+    if (timestamp - msg_timestamp) > timeout:
+        raise DecryptionTimeout("The message has an expired timeout")
+    return message.decode("utf-8")
+
+
+def handle_decrypt_message(args):
+    key = base64.b64decode(args.key.encode("utf-8"))
+    ciphertext = base64.b64decode(args.message.encode("utf-8"))
+    message = decrypt_message(key, ciphertext, timeout=args.timeout)
+    print(str(message), file=args.outfile)
+
+
+def parse_args(argv=None):
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument("--outfile", default=sys.stdout, type=argparse.FileType("w"), help="The output file")
+
+    subparsers = parser.add_subparsers(help="Available sub-commands")
+
+    gen_parser = subparsers.add_parser("generate_key", help="Generate a random key string")
+    gen_parser.set_defaults(func=handle_generate_key)
+
+    encrypt_parser = subparsers.add_parser("encrypt", help="Encrypt a message")
+    encrypt_parser.add_argument("--key", required=True, help="The encryption key")
+    encrypt_parser.add_argument("--message", required=True, help="Message to encrypt")
+    encrypt_parser.set_defaults(func=handle_encrypt_message)
+
+    decrypt_parser = subparsers.add_parser("decrypt", help="Decyrpt a message")
+    decrypt_parser.add_argument("--key", required=True, help="The encryption key")
+    decrypt_parser.add_argument("--message", required=True, help="Message to decrypt")
+    decrypt_parser.add_argument(
+        "--timeout",
+        default=MESSAGE_TIMEOUT,
+        type=int,
+        help="A message timeout. Decryption will fail for older messages",
+    )
+    decrypt_parser.set_defaults(func=handle_decrypt_message)
+
+    return parser.parse_args(argv)
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    args.func(args)

.github/workflows/nf-test.yml

@@ -97,6 +97,10 @@ jobs:
         continue-on-error: ${{ matrix.NXF_VER == 'latest-everything' }}
         env:
           NFT_WORKDIR: ${{ env.NFT_WORKDIR }}
+          SENTIEON_AUTH_MECH: "GitHub Actions - token"
+          SENTIEON_ENCRYPTION_KEY: ${{ secrets.SENTIEON_ENCRYPTION_KEY }}
+          SENTIEON_LICENSE_MESSAGE: ${{ secrets.SENTIEON_LICENSE_MESSAGE }}
+          SENTIEON_LICSRVR_IP: ${{ secrets.SENTIEON_LICSRVR_IP }}
         with:
           profile: ${{ matrix.profile }}
           shard: ${{ matrix.shard }}

CHANGELOG.md

@@ -11,6 +11,7 @@ Special thanks to the following for their contributions to the release:
 
 - [Jonathan Manning](https://github.com/pinin4fjords)
 - [Maxime Garcia](https://github.com/maxulysse)
+- [Friederike Hanssen](https://github.com/friederikehanssen)
 - [Ido Tamir](https://github.com/idot)
 - [Usman Rashid](https://github.com/GallVp)
 
@@ -21,12 +22,14 @@ Special thanks to the following for their contributions to the release:
 - [PR #1573](https://github.com/nf-core/rnaseq/pull/1573) - Fix salmon.merged.SummarizedExperiment.rds name collision
 - [PR #1585](https://github.com/nf-core/rnaseq/pull/1585) - Update awsfulltest.yml to restore aligner-wise outputs
 - [PR #1580](https://github.com/nf-core/rnaseq/pull/1580) - Template update for nf-core/tools v3.3.2
+- [PR #1590](https://github.com/nf-core/rnaseq/pull/1590) - Addition of Sentieon STAR
 
 ### Software dependencies
 
 | Dependency | Old version | New version |
 | ---------- | ----------- | ----------- |
 | `MultiQC`  | 1.29        | 1.30        |
+| `Sentieon` |             | 202503.01   |
 
 ## [[3.19.0](https://github.com/nf-core/rnaseq/releases/tag/3.19.0)] - 2025-06-10
 

README.md

@@ -33,7 +33,7 @@
 5. Adapter and quality trimming ([`Trim Galore!`](https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/))
 6. Removal of genome contaminants ([`BBSplit`](http://seqanswers.com/forums/showthread.php?t=41288))
 7. Removal of ribosomal RNA ([`SortMeRNA`](https://github.com/biocore/sortmerna))
-8. Choice of multiple alignment and quantification routes:
+8. Choice of multiple alignment and quantification routes (_For `STAR` the sentieon implementation can be chosen_):
    1. [`STAR`](https://github.com/alexdobin/STAR) -> [`Salmon`](https://combine-lab.github.io/salmon/)
    2. [`STAR`](https://github.com/alexdobin/STAR) -> [`RSEM`](https://github.com/deweylab/RSEM)
    3. [`HiSAT2`](https://ccb.jhu.edu/software/hisat2/index.shtml) -> **NO QUANTIFICATION**

conf/test.config

@@ -45,7 +45,7 @@ params {
 
 // When using RSEM, remove warning from STAR whilst building tiny indices
 process {
-    withName: 'RSEM_PREPAREREFERENCE_GENOME' {
+    withName: 'RSEM_PREPAREREFERENCE_GENOME|SENTIEON_RSEMPREPAREREFERENCE_GENOME' {
         ext.args2 = "--genomeSAindexNbases 7"
     }
 }

docs/usage.md

@@ -139,6 +139,34 @@ You can use `--skip_alignment --skip_pseudo_alignment` if you only want to run t
 
 Note that `--skip_alignment` and `--skip_pseudo_alignment` prevent both the execution of alignment/pseudoalignment steps and the building of their corresponding indices. For example, using `--skip_alignment` with `--aligner star_salmon` will skip both STAR alignment and index building.
 
+### Sentieon acceleration for STAR
+
+The STAR aligner can be accelerated through its Sentieon implemention using the parameter `--use_sentieon_star`.
+
+Sentieon is a commercial solution to process genomics data, requiring a paid license. Sentieon's tooling contains an accelerated version of the [`STAR` aligner](https://support.sentieon.com/manual/usages/general/?highlight=star#star-binary), which nf-core/rnaseq supports. In order to use those functions, the user will need to supply a license for Sentieon.
+
+Sentieon supply license in the form of a string-value (a url) or a file. It should be base64-encoded and stored in a nextflow secret named `SENTIEON_LICENSE_BASE64`. If a license string (url) is supplied, then the nextflow secret should be set like this:
+
+```bash
+nextflow secrets set SENTIEON_LICENSE_BASE64 $(echo -n <sentieon_license_string> | base64 -w 0)
+```
+
+:::note
+<sentieon_license_string> is formatted as `IP:Port` for example: `12.12.12.12:8990`
+:::
+
+If a license file is supplied, then the nextflow secret should be set like this:
+
+```bash
+nextflow secrets set SENTIEON_LICENSE_BASE64 \$(cat <sentieon_license_file.lic> | base64 -w 0)
+```
+
+:::note
+If you're looking for documentation on how the nf-core Sentieon GitHub Actions and Sentieon License Server are set up: [Here be dragons.](https://github.com/nf-core/ops/blob/main/pulumi/sentieon_license_server/README.md)
+
+For detailed instructions on how to test the modules and subworkflows separately, see [here](https://github.com/nf-core/modules/blob/master/modules/nf-core/sentieon/README.md).
+:::
+
 ## Quantification options
 
 The current options align with STAR and quantify using either Salmon (`--aligner star_salmon`) / RSEM (`--aligner star_rsem`). You also have the option to pseudoalign and quantify your data with Salmon or Kallisto by providing the `--pseudo_aligner salmon` or `--pseudo_aligner kallisto` parameter, respectively.

main.nf

@@ -84,7 +84,8 @@ workflow NFCORE_RNASEQ {
         params.skip_bbsplit,
         !params.remove_ribo_rna,
         params.skip_alignment,
-        params.skip_pseudo_alignment
+        params.skip_pseudo_alignment,
+        params.use_sentieon_star
     )
     ch_versions = ch_versions.mix(PREPARE_GENOME.out.versions)
 

modules.json

@@ -217,6 +217,21 @@
                         "git_sha": "05954dab2ff481bcb999f24455da29a5828af08d",
                         "installed_by": ["bam_stats_samtools"]
                     },
+                    "sentieon/rsemcalculateexpression": {
+                        "branch": "master",
+                        "git_sha": "2779d18605e9923332155d671f45ed37fa185ff4",
+                        "installed_by": ["modules"]
+                    },
+                    "sentieon/rsempreparereference": {
+                        "branch": "master",
+                        "git_sha": "2779d18605e9923332155d671f45ed37fa185ff4",
+                        "installed_by": ["modules"]
+                    },
+                    "sentieon/staralign": {
+                        "branch": "master",
+                        "git_sha": "73d3ab1ac411843ae643b5bc310a4aec9ae7ab3a",
+                        "installed_by": ["modules"]
+                    },
                     "sortmerna": {
                         "branch": "master",
                         "git_sha": "d4a425ce59fc803a11e520f16680ede9af09761f",