Data

Tools

Indexes

Applications

Experiments

Open Source Science

Recent Releases of d-script

d-script - v0.3.1: 2025-08-12 -- Update documentation and biotite use

What's Changed

  • Update doc building and testing by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/79
  • Update to use biotite instead of biopython and standalone Foldseek by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/80

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.3.0...v0.3.1

- Python
Published by samsledje 11 months ago

d-script - v0.3.0: 2025-07-22 -- Blocked, Multi-GPU Parallel Inference (BMPI)

What's Changed

  • Update evaluate.py and predict.py by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/76
  • Modernization of D-SCRIPT in anticipation of v0.3.0 by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/77
  • Update README.md by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/78
  • Parallel, multi-GPU pairwise inference by @schafferde in https://github.com/samsledje/D-SCRIPT/pull/75

New Contributors

  • @schafferde made their first contribution in https://github.com/samsledje/D-SCRIPT/pull/75

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.2.8...v0.3.0

- Python
Published by samsledje 12 months ago

d-script - v0.2.8: 2024-09-16 -- Fix bug with model cuda assignment during prediction

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.2.7...v0.2.8

- Python
Published by samsledje almost 2 years ago

d-script - v0.2.7: 2024-09-13 -- HuggingFace Models

What's Changed

  • R notebooks added for baseline sequence-based analysis by @kap-devkota in https://github.com/samsledje/D-SCRIPT/pull/59
  • Add HuggingFace support by @samsledje in https://github.com/samsledje/D-SCRIPT/pull/69

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.2.6...v0.2.7

- Python
Published by samsledje almost 2 years ago

d-script - v0.2.6b: 2023-08-24 -- Zenodo Release

Release to generate a DOI for the current state of the repository containing TT3D experiments.

What's Changed

  • R notebooks added for baseline sequence-based analysis by @kap-devkota in https://github.com/samsledje/D-SCRIPT/pull/59
  • Python notebooks for baseline structure-based analysis

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.2.6...v0.2.6b

- Python
Published by samsledje over 2 years ago

d-script - v0.2.6: 2023-09-28 -- Add TT3D as a pretrained option

Full Changelog: https://github.com/samsledje/D-SCRIPT/compare/v0.2.5...v0.2.6

- Python
Published by samsledje almost 3 years ago

d-script - v0.2.5: 2023-07-24 -- Integrating 3Di sequences into prediction

Update predict.py so models trained with --allow_foldseek (Topsy-Turvy 3D) can be used to make predictions, provided that a foldseek_fasta is provided.

- Python
Published by samsledje almost 3 years ago

d-script - v0.2.4

TT3D -- Integrating FoldSeek Structures into Topsy-Turvy training

- Python
Published by samsledje about 3 years ago

d-script - v0.2.2: 2023-01-20 -- Bug Fixes and Topsy-Turvy

  • Resolve #35 to use require_dataset -- can now add multiple .fasta files to the same h5 file
  • Update pretrained API and docs to include Topsy-Turvy
  • Add retry decorator to get_pretrained if download fails
  • Add ability to set a random seed for training
  • Update evaluate code to also store metrics in a file

- Python
Published by samsledje over 3 years ago

d-script - v0.2.1: 2022-06-28 -- Bug Fixes

  • Update setup.py to include biopython so that it runs properly on a clean pip install.

- Python
Published by samsledje about 4 years ago

d-script - v0.2.0: 2022-06-24 -- Integration of Topsy-Turvy

Topsy-Turvy Integration

  • Integrate Topsy-Turvy to allow for top-down supervision
  • See the GitHub repository
  • See the landing page
  • To appear at ISMB 2022 and in Bioinformatics as Devkota*, Singh*, Sledzieski, Berger, Cowen, “Topsy-Turvy: integrating a global view into sequence-based PPI prediction”

Bug Fixes and Quality of Life

  • Use utils.log function across all commands
  • Speed up loading embeddings into memory using parallel processing
  • Update fasta.parse and fasta.write to use BioPython SeqIO (better error checking)
  • More comprehensive test suite for main commands

- Python
Published by samsledje about 4 years ago

d-script - v0.1.9 Dealing with pypi and semvar fussiness

This is the same as v0.1.8b0 but it will actually take priority

- Python
Published by samsledje over 4 years ago

d-script - v0.1.8b0: 2022-03-07 -- Minor bug fixes

  • Updated model loading on new version to handle re-named parameters
  • Updated cpu-only loading during prediction with map_location

- Python
Published by samsledje over 4 years ago

d-script - v0.1.8: 2022-02-08 -- Finalized training bug fix

  • Resolve #24 by fixing training
  • Can now run dscript train --train data/pairs/human_train.tsv --test data/pairs/human_test.tsv --embedding /afs/csail/u/s/samsl/Work/databases/STRING/homo.sapiens/human_nonRed. h5 --output [output] --save-prefix [prefix] --device 0 to replicate paper results
  • Updated code formatting with black and pre-commit

- Python
Published by samsledje over 4 years ago

d-script - Bug Fix - Augmentation and proper defaults

  • Augmentation fix in v0.1.5 was bugged still and would throw an error, now resets index
  • Change --use-w and --augment to --no-w and --no-augment with store false

- Python
Published by samsledje almost 5 years ago

d-script - Bug Fix - Augment and Documentation

v0.1.5: 2021-06-23 -- Bug Fix - Augment and Documentation

  • Updated package level imports
  • Updated documentation
  • Fixed issue #13: improper augmentation of data
  • Fixed issue #12: overwrites cmap data sets if they already exist

- Python
Published by samsledje almost 5 years ago

d-script - Bug Fix - Typo in ContactModule.forward()

Fixed Issue #7 - bug which would crash contact module if called directly.

Release for Zenodo.

- Python
Published by samsledje over 5 years ago

d-script - Bug Fix - Pairs too large for GPU

  • Fixed issues #3, #4 (probably didn't need to open two issues! didn't realize one was already open)!
  • Basic logging system implemented to report skipped pairs
  • Fixed wrong variable name in loading from sequence file
  • Updated documentation

- Python
Published by samsledje over 5 years ago

d-script - Bug Fix - Eval Mode

Model should be put into eval() mode before prediction or evaluation, and when new models are downloaded - this makes the output deterministic by disabling dropout layers.

- Python
Published by samsledje over 5 years ago

d-script - First Beta Release

See documentation.

- Python
Published by samsledje over 5 years ago