SPOT-1D-Single
Model checkpoints for SPOT-1D-Single, which predicts protein secondary structure (3- and 8-state), solvent accessibility (ASA), half-sphere exposure (HSE-up and HSE-down), contact number and backbone angles (ฯ, ฯ, ฮธ, ฯ) from a single sequence, without multiple sequence alignments or evolutionary profiles.
Code and instructions: https://github.com/jas-preet/SPOT-1D-Single
Files
| File | Size | SHA-256 |
|---|---|---|
jits.tar.xz |
1.2 GB | 8fef1e8091866f43f56f6c9e02f719607e0825a64c37d7fe8df7c2e3a07d1ea0 |
The archive holds TorchScript checkpoints for an ensemble of three models (a bidirectional LSTM, a multi-scale ResNet and a multi-scale ResNet-LSTM), for classification and for regression, in CPU and GPU versions.
git clone https://github.com/jas-preet/SPOT-1D-Single.git
cd SPOT-1D-Single
curl -LO https://huggingface.co/Jas-preet/SPOT-1D-Single/resolve/main/jits.tar.xz
echo "8fef1e8091866f43f56f6c9e02f719607e0825a64c37d7fe8df7c2e3a07d1ea0 jits.tar.xz" | shasum -a 256 -c
tar -xf jits.tar.xz
Training data
Trained on 39,120 proteins from ProteinNet (PDB entries before 2016, resolution < 2.5 ร , clustered at 95% sequence identity). See the paper for the independent test sets (TEST2018, SPOT-2016, SPOT-2018, CASP12-FM, CASP13-FM).
Citation
Singh J, Litfin T, Paliwal K, Singh J, Hanumanthappa AK, Zhou Y. SPOT-1D-Single: improving the single-sequence-based prediction of protein secondary structure, backbone angles, solvent accessibility and half-sphere exposures using a large training set and ensembled deep learning. Bioinformatics. 2021;37(20):3464-3472. doi:10.1093/bioinformatics/btab316