HAL

Hexamer Additive Linear model for predicting alternative splicing from sequence.

Disclaimer

This is an UNOFFICIAL implementation of Learning the Sequence Determinants of Alternative Splicing from Millions of Random Sequences by Alexander B. Rosenberg, Rupali P. Patwardhan, Jay Shendure and Georg Seelig.

The OFFICIAL repository of HAL is at Alex-Rosenberg/cell-2015.

The MultiMolecule team has confirmed that the provided model and checkpoints are producing the same intermediate representations as the original implementation.

The team releasing HAL did not write this model card for this model so this model card has been written by the MultiMolecule team.

Model Details

HAL is a linear (additive) model that scores alternative 5' splice-site usage from normalized hexamer (6-mer) frequencies across a 160-nucleotide donor-region window. It was learned from massively parallel reporter assays measuring splicing of millions of random synthetic sequences. The published checkpoint contains a (4096, 8) matrix of hexamer coefficients; MultiMolecule follows the upstream HAL formula by averaging the eight coefficient columns into one effect per hexamer and applying those effects to normalized hexamer frequencies.

Model Specification

Window	Published Coefficient Columns	Hexamer Features	Num Parameters	FLOPs	MACs
160 nt	8 averaged	4,096	4,096	8,192	4,096

• Model type: linear model over normalized hexamer (6-mer) frequency features
• k-mer size: 6
• Input window: 160 nucleotides
• Number of hexamer features: 4,096

Links

• Code: multimolecule.hal
• Paper: Learning the Sequence Determinants of Alternative Splicing from Millions of Random Sequences
• Developed by: Alexander B. Rosenberg, Rupali P. Patwardhan, Jay Shendure, Georg Seelig
• Original Repository: Alex-Rosenberg/cell-2015

Usage

The model file depends on the multimolecule library. You can install it using pip:

pip install multimolecule

Direct Use

Alternative Splicing Prediction

You can use this model directly to predict a splicing score for a 160-nucleotide DNA sequence window:

>>> import torch
>>> from multimolecule import DnaTokenizer, HalForSequencePrediction

>>> tokenizer = DnaTokenizer.from_pretrained("multimolecule/hal")
>>> model = HalForSequencePrediction.from_pretrained("multimolecule/hal")
>>> sequence = "ACGT" * 40
>>> input = tokenizer(sequence, add_special_tokens=False, return_tensors="pt")
>>> output = model(**input)

>>> output.logits.shape
torch.Size([1, 1])

Training Details

HAL was learned from massively parallel splicing reporter assays in which millions of random synthetic sequences were inserted into an alternatively spliced reporter minigene. Splicing outcomes were measured by high-throughput sequencing of the resulting mRNA isoforms.

Training Data

The model was trained on the splicing measurements of millions of degenerate (random) sequences from the reporter library described in the HAL paper. Hexamer coefficients were estimated by regressing the measured splicing index against the hexamer composition of each sequence.

Training Procedure

Pre-training

HAL is a linear regression model. The published hexamer coefficient table is fit to the measured splicing index, and the model prediction is the linear combination of normalized hexamer frequencies with the averaged hexamer effects.

Conversion And Provenance

• The checkpoint is converted from the original HAL_mer_scores.npz artifact referenced by the official Alex-Rosenberg/cell-2015 repository and the Kipoi HAL model.
• The original artifact stores 4,096 hexamer rows and eight coefficient columns. MultiMolecule averages those columns into the single per-hexamer effect used by the upstream HAL formula.

Citation

@article{rosenberg2015learning,
  author    = {Rosenberg, Alexander B. and Patwardhan, Rupali P. and Shendure, Jay and Seelig, Georg},
  journal   = {Cell},
  number    = 3,
  pages     = {698--711},
  publisher = {Elsevier BV},
  title     = {Learning the Sequence Determinants of Alternative Splicing from Millions of Random Sequences},
  volume    = 163,
  year      = 2015,
  doi       = {10.1016/j.cell.2015.09.054}
}

The artifacts distributed in this repository are part of the MultiMolecule project. If you use MultiMolecule in your research, you must cite the MultiMolecule project as follows:

@software{chen_2024_12638419,
  author    = {Chen, Zhiyuan and Zhu, Sophia Y.},
  title     = {MultiMolecule},
  doi       = {10.5281/zenodo.12638419},
  publisher = {Zenodo},
  url       = {https://doi.org/10.5281/zenodo.12638419},
  year      = 2024,
  month     = may,
  day       = 4
}

Known Limitations

• HAL scores a single fixed 160-nucleotide donor-region window. Paired donor comparisons and sigmoid postprocessing can be built by subtracting two window scores.
• HAL uses the four canonical DNA nucleotides ACGT. Any hexamer spanning an unknown / N token is ignored.
• HAL is a linear hexamer model; it does not capture long-range splicing signals beyond the published 160-nucleotide window.

Contact

Please use GitHub issues of MultiMolecule for any questions or comments on the model card.

Please contact the authors of the HAL paper for questions or comments on the paper/model.

License

This model implementation is licensed under the GNU Affero General Public License.

For additional terms and clarifications, please refer to our License FAQ.

SPDX-License-Identifier: AGPL-3.0-or-later