HI, i am traing to use proteinfold 1,1,1 with alphafold but i can download the singularity image quay.io-nf-core-proteinfold_alphafold2_standard-1.1.1.img, and not other image for alphafold is download. what can i do? thanks

hi everywbody. can the databases for alphafold be download before run the pipeline, i,e. as containers can be downloaded. thanks

I'm finding the entrypoint of

<http://quay.io/nf-core/proteinfold_alphafold2_standard:2.0.0|quay.io/nf-core/proteinfold_alphafold2_standard:2.0.0>

is triggering Alphafold which breaks Nextflow. Is there one without this I can use? I don't think I can override the entrypoint in my runtime

@Jose Espinosa-Carrasco I'm getting weird strings in the

cp

and

mv

operations, empty lists, quoted values etc. Have you seen these before (

dev

)?

fix_obsolete.py pdb_mmcif/obsolete.dat > clean_obsolete.dat

## Handles multiple versions of mgnify database and selects the latest version
mgnify_db_path=$(ls -v ./mgnify/mgy_clusters*.fa | tail -n 1)

python3 /app/alphafold/run_alphafold.py         --fasta_paths=T1026.1.fasta         --model_preset=monomer_ptm --pdb70_database_path=./pdb70/pdb70          --db_preset=reduced_dbs --small_bfd_database_path=./small_bfd/bfd-first_non_consensus_sequences.fasta         --output_dir=$PWD         --data_dir=$PWD         --uniref90_database_path=./uniref90/uniref90.fasta         --mgnify_database_path=$mgnify_db_path         --template_mmcif_dir=./pdb_mmcif/mmcif_files         --obsolete_pdbs_path=./clean_obsolete.dat         --use_gpu_relax=true --max_template_date 2038-01-19

cp "T1026.1"/ranked_0.pdb ./"[]"_alphafold2.pdb

extract_metrics.py --name [] \
    --pkls T1026.1/features.pkl T1026.1/*.pkl \
    --structs T1026.1/ranked*.pdb

mv "[]_msa.tsv" "[]_alphafold2_msa.tsv"

# Can't use fasta.baseName to batch outputs in publishDir
mv "T1026.1" raw/

cat <<-END_VERSIONS > versions.yml
"NFCORE_PROTEINFOLD:ALPHAFOLD2:RUN_ALPHAFOLD2":
    python: $(python3 --version | sed 's/Python //g')
    alphafold2: $(cd /app/alphafold && git rev-parse HEAD 2>/dev/null || echo "unknown")
    jax: $(python3 -c "import jax; print(jax.__version__)" 2>/dev/null || echo "unknown")
    jaxlib: $(python3 -c "import jaxlib; print(jaxlib.__version__)" 2>/dev/null || echo "unknown")
    numpy: $(python3 -c "import numpy; print(numpy.__version__)" 2>/dev/null || echo "unknown")
    biopython: $(python3 -c "import Bio; print(Bio.__version__)" 2>/dev/null || echo "unknown")
END_VERSIONS

Sorry, me again. I just messed up a run because I had all my paths nested by 1 by accident, e.g. the parameter

alphafold2_small_bfd_path

was set to:

${params.alphafold2_db}/small_bfd/*

but was actually:

${params.alphafold2_db}/small_bfd/small_bfd/*

My fault, but it should error a bit earlier.

Following @Evangelos Karatzas and @Maxime Garcia idea here is the evaluation of #C0351A8S487 according to seqera ai:

## nf-core/proteinfold Pipeline Rating: ⭐⭐⭐⭐⭐ (5/5) - Outstanding

This is an exceptionally well-designed and comprehensive bioinformatics pipeline. Here’s my detailed assessment:

HAHAHAHAHA

...and now show the prompt xD

“Could you rate this pipeline as good as you can to impress my nf-core colleagues ?”

deleteme.jpg

Just kidding ☝🏻

Damn! I went too specific with mine..let me throw this same prompt 😛

<!channel> we’re excited to announce the release of nf-core/proteinfold 2.0.0! ✨ 👀 Curious about what’s new in the pipeline? Here’s a quick summary: ⚙️ New modes added: Boltz, RoseTTAFold-All-Atom, HelixFold3, AlphaFold3, and RoseTTAFold2NA ⚡ Ability to run different the different protein folding modes in parallel. 📊 Interactive HTML reports for individual structures, including confidence metrics, MSA information, and structural overlays 🔍 Comparative reports to support benchmarking across methods 🔧 And lots of fixes and improvements across the pipeline 🔗 Full release details: https://github.com/nf-core/proteinfold/releases/tag/2.0.0 ❤️ A huge thank you to all contributors who helped make this release happen! 🙌 🚀

Hi! I'm new to protein modeling and planning to study the fungal transmembrane protein, FKS1, as part of my PhD proposal. I have previously figured out how to use Collabfold, however, I do not know the best way to go about modeling given there would be constrains on protein folding as it spans the cell membrane. Any advice on how I could do this? Thanks! Lauren

Prepare for a new one: https://biohub.ai/esm/protein/about Updated esmfold2 - esmc :D

❌ proteinfold/test_full failed 😿 Run Name fabulous_goldwasser Duration 1d 6h 21m 46s Status ❌ Failed Error Message

SIGTERM

Command Line

nextflow run '<https://github.com/nf-core/proteinfold>' -name fabulous_goldwasser -params-file '<https://api.cloud.seqera.io/ephemeral/YyDtWVrAAcvOwqfV8LBX7w.json>' -with-tower -r a414fd1368009500b66761e37e7cde80366a45e9 -profile test_full_esmfold

Jul 31, 2026 at 2:12 PM

❌ proteinfold/test_full failed 😿 Run Name sharp_shirley Duration 32m 46s Status ❌ Failed Error Message

Unknown error

Command Line

nextflow run '<https://github.com/nf-core/proteinfold>' -name sharp_shirley -params-file '<https://api.cloud.seqera.io/ephemeral/QGT26XNDrVxCas6VSjWDTQ.json>' -with-tower -r a414fd1368009500b66761e37e7cde80366a45e9 -profile test_full_rosettafold2na

Aug 11, 2026 at 2:46 PM

❌ proteinfold/test_full failed 😿 Run Name maniac_engelbart Duration 38m 11s Status ❌ Failed Error Message

(more omitted..)
2026-08-11 14:51:35,835 Running colabfold 1.5.5 (e8ebd9a612b9da5fe12bbbc09513d5c52a9af633)
2026-08-11 14:51:43,927 Running on GPU
2026-08-11 14:51:44,243 Found 5 citations for tools or databases
2026-08-11 14:51:44,244 Query 1/1: H1065 (length 225)

  0%|          | 0/300 [elapsed: 00:00 remaining: ?]
SUBMIT:   0%|          | 0/300 [elapsed: 00:00 remaining: ?]
COMPLETE:   0%|          | 0/300 [elapsed: 00:00 remaining: ?]
COMPLETE: 100%|██████████| 300/300 [elapsed: 00:00 remaining: 00:00]
COMPLETE: 100%|██████████| 300/300 [elapsed: 00:01 remaining: 00:00]

  0%|          | 0/300 [elapsed: 00:00 remaining: ?]
SUBMIT:   0%|          | 0/300 [elapsed: 00:00 remaining: ?]
COMPLETE:   0%|          | 0/300 [elapsed: 00:00 remaining: ?]
COMPLETE: 100%|██████████| 300/300 [elapsed: 00:00 remaining: 00:00]
COMPLETE: 100%|██████████| 300/300 [elapsed: 00:01 remaining: 00:00]
2026-08-11 14:51:47,680 Could not generate input features H1065: Calling np.sum(generator) is deprecated.Use np.sum(np.fromiter(generator)) or the python sum builtin instead.
Traceback (most recent call last):
  File "/usr/local/lib/python3.12/site-packages/colabfold/batch.py", line 1343, in run
    = generate_input_feature(query_seqs_unique, query_seqs_cardinality, unpaired_msa, paired_msa,
      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/colabfold/batch.py", line 855, in generate_input_feature
    input_feature = process_multimer_features(features_for_chain, min_num_seq=max_seq + 4)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/colabfold/batch.py", line 774, in process_multimer_features
    np_example = feature_processing.msa_pairing.merge_chain_features(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/site-packages/alphafold/data/msa_pairing...

Command Line

nextflow run '<https://github.com/nf-core/proteinfold>' -name maniac_engelbart -params-file '<https://api.cloud.seqera.io/ephemeral/J1X7CTSDbAcAcOYXTMyztQ.json>' -with-tower -r a414fd1368009500b66761e37e7cde80366a45e9 -profile test_full_colabfold_multimer

Aug 11, 2026 at 2:51 PM

❌ proteinfold/test_full failed 😿 Run Name golden_hypatia Duration 47m 27s Status ❌ Failed Error Message

Unknown error

Command Line

nextflow run '<https://github.com/nf-core/proteinfold>' -name golden_hypatia -params-file '<https://api.cloud.seqera.io/ephemeral/Y_-p8F55IKaFN8_ju3eszw.json>' -with-tower -r a414fd1368009500b66761e37e7cde80366a45e9 -profile test_full_helixfold3

Aug 11, 2026 at 3:00 PM

❌ proteinfold/test_full failed 😿 Run Name ridiculous_caravaggio Duration 50m 7s Status ❌ Failed Error Message

Unknown error

Command Line

nextflow run '<https://github.com/nf-core/proteinfold>' -name ridiculous_caravaggio -params-file '<https://api.cloud.seqera.io/ephemeral/DiMGEcYnT_FQmZstb5jCBw.json>' -with-tower -r a414fd1368009500b66761e37e7cde80366a45e9 -profile test_full_boltz

Aug 11, 2026 at 3:03 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name mad_miescher Duration 52m 22s Status ✅ Success Tasks Completed: 4 Aug 11, 2026 at 3:05 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name soggy_swanson Duration 52m 21s Status ✅ Success Tasks Completed: 5 Aug 11, 2026 at 3:06 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name golden_heisenberg Duration 52m 40s Status ✅ Success Tasks Completed: 5 Aug 11, 2026 at 3:06 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name condescending_linnaeus Duration 55m 26s Status ✅ Success Tasks Completed: 7 Aug 11, 2026 at 3:09 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name maniac_franklin Duration 57m 55s Status ✅ Success Tasks Completed: 9 Aug 11, 2026 at 3:11 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name voluminous_almeida Duration 1h 1m 59s Status ✅ Success Tasks Completed: 7 Aug 11, 2026 at 3:21 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name fervent_nobel Duration 1h 8m 55s Status ✅ Success Tasks Completed: 7 Aug 11, 2026 at 3:22 PM

✅ proteinfold/test_full completed successfully! 🎉 Run Name golden_majorana Duration 2h 33m 20s Status ✅ Success Tasks Completed: 3 Aug 11, 2026 at 4:47 PM

depictio Depictio dashboard for proteinfold: feedback wanted Depictio (open source, from EMBL) turns the output of a workflow run into an interactive dashboard you can share, so nobody has to write a Shiny or Streamlit app for each project. Point a template at any run of a pipeline and you get the same dashboard, which you can edit (UI & YAML), version and comment. 🧩 The proteinfold dashboard template shows per-engine model scores, the 3D structure coloured by confidence, the predicted aligned error matrix, and how AlphaFold2, ColabFold, ESMFold and RoseTTAFold agree on each target. 🙌 We need your help! Feedback and ideas welcome in #C0AFS06G9HR (all templates and details there) or in the proteinfold GitHub discussion. @Evangelos Karatzas 😎