Biocentral Documentation

Biocentral Documentation

Modules

Embed: Get meaningful representations of your data

Train: Train your own machine learning models from scratch

Predict: Use pretrained models to predict protein properties

TMbed LightAttentionMembrane LightAttentionSubcellularLocalization Seth UdonPred BindEmbed ProtT5Conservation ProtT5SecondaryStructure VespaG ExoTox

Active Learning: Optimize your lab processes

Proteins: Useful tools for the backbone of life

Predict: Use pretrained models to predict protein properties

Seth

Residue-level disorder scoring with CheZOD-compatible interpretation.

Summary

Model name: Seth
Task: Predict nuances of residue disorder in proteins
Protocol: residue_to_value
Embedder: Rostlab/prot_t5_xl_uniref50
Model size: 575.1 KB

Outputs

disorder_chezod

Description: Disorder scores based on CheZOD Z-score interpretation
Output type: per_residue
Value type: float
Interpretation: Values below 8 indicate disorder; values above 8 indicate order

References

Authors: Dagmar Ilzhoefer; Michael Heinzinger; Burkhard Rost
Model link: github.com/DagmarIlz/SETH
Citation: doi.org/10.1101/2022.06.23.497276
License: GPL-3.0
Training data: data.bioembeddings.com/public/design

LightAttentionSubcellularLocalization

Sequence-level prediction of protein subcellular localization classes.

UdonPred

Residue-level intrinsic disorder scoring using TriZOD outputs.

On this page

Summary Outputs disorder_chezod References