use crate::{
LaunchType, License, LicenseCategory, ProcessExpense, SpecData, ToolCategory,
tool_definitions::{
Tool,
catalog::{CatalogEntry, DataType, Identity, PrimaryInput},
},
};
pub const ENTRY: CatalogEntry = CatalogEntry {
identity: Identity::Installed(Tool::HighFold),
categories: &[
ToolCategory::StructurePrediction,
ToolCategory::PeptideBinderDesign,
],
launch_type: LaunchType::CondaBasedApp,
license_type: LicenseCategory::Permissive,
expense: ProcessExpense::Expensive,
primary_output: Some(DataType::Pdb),
primary_inputs: &[PrimaryInput::residues(
"peptide_sequence",
&[DataType::AaSequence],
)],
top_choice: false,
spec: SpecData {
summary: "Predict the structure of a cyclic peptide or a cyclic-peptide complex.",
description: "Runs HighFold, which folds cyclic peptides by feeding AlphaFold 2 a cyclic position offset matrix in place of its usual relative positional encoding, so head-to-tail and disulfide-bridged macrocycles are modelled as closed rather than as linear chains whose ends happen to meet.",
availability: "Conda environment, an NVIDIA GPU, and the AlphaFold 2 parameters are required",
license_details: "HighFold modifies ColabFold (MIT) and runs on the AlphaFold 2 parameters, which DeepMind publishes under CC BY 4.0. Both permit commercial use; confirm the upstream repository's own terms before relying on that.",
repo_url: Some("https://github.com/hongliangduan/HighFold"),
home_url: None,
docs_url: Some("https://academic.oup.com/bib/article/25/3/bbae215/7665139"),
input_params_url: None,
examples_url: None,
paper_url: Some("https://academic.oup.com/bib/article/25/3/bbae215/7665139"),
license: License::Mit,
license_url: None,
tested: false,
},
};