KOFFI and Anabel 2.0—a new binding kinetics database and its integration in an open-source binding analysis software

Norval, Leo; Krämer, Stefan; Gao, Mingjie; Herz, Tobias; Li, Jianyu; Rath, Christin; Wöhrle, Johannes; Günther, Stefan; Roth, Günter

doi:10.1093/database/baz101

Cited by 10 publications

(13 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…KOFFI is a similar database. 19 It added more information, such as the assay method, device, chip, and so on. It collects a total of 1705 individual entries, most of which are kinetic data of protein–protein and protein–nucleic acid complexes, with very few records of protein–small molecules.…”

Section: Resultsmentioning

confidence: 99%

Public Data Set of Protein–Ligand Dissociation Kinetic Constants for Quantitative Structure–Kinetics Relationship Studies

Liu

Lin

et al. 2022

ACS Omega

View full text Add to dashboard Cite

Protein–ligand binding affinity reflects the equilibrium thermodynamics of the protein–ligand binding process. Binding/unbinding kinetics is the other side of the coin. Computational models for interpreting the quantitative structure–kinetics relationship (QSKR) aim at predicting protein–ligand binding/unbinding kinetics based on protein structure, ligand structure, or their complex structure, which in principle can provide a more rational basis for structure-based drug design. Thus far, most of the public data sets used for deriving such QSKR models are rather limited in sample size and structural diversity. To tackle this problem, we have compiled a set of 680 protein–ligand complexes with experimental dissociation rate constants ( k off ), which were mainly curated from the references accumulated for updating our PDBbind database. Three-dimensional structure of each protein–ligand complex in this data set was either retrieved from the Protein Data Bank or carefully modeled based on a proper template. The entire data set covers 155 types of protein, with their dissociation kinetic constants ( k off ) spanning nearly 10 orders of magnitude. To the best of our knowledge, this data set is the largest of its kind reported publicly. Utilizing this data set, we derived a random forest (RF) model based on protein–ligand atom pair descriptors for predicting k off values. We also demonstrated that utilizing modeled structures as additional training samples will benefit the model performance. The RF model with mixed structures can serve as a baseline for testifying other more sophisticated QSKR models. The whole data set, namely, PDBbind-koff-2020 , is available for free download at our PDBbind-CN web site ( ).

show abstract

Section: Resultsmentioning

confidence: 99%

Public Data Set of Protein–Ligand Dissociation Kinetic Constants for Quantitative Structure–Kinetics Relationship Studies

Liu

Lin

et al. 2022

ACS Omega

View full text Add to dashboard Cite

show abstract

“…Targeted proteins can easily be engineered to be highly selective and have affinities that are considered optimal by our modeling approach ( Figure 3 A). However, for peptides, the ability to obtain high affinities appears to be a limiting factor for their efficacy ( Figure 3 B), although peptides with very high affinities have been reported [ 30 ], including peptides that bind with subnanomolar K d values to vascular endothelial growth factor (VEGF) receptor 2 and c-MET, or hapten peptides that bind to single chain variable fragments with affinities as high as 2.3 nM [ 31 ]. Interestingly, for targeted proteins, the optimal affinities for targeting tumors with and without convection, of 16.4 nM and 625 pM, respectively, for these simulated parameters, appear to be readily achievable, as demonstrated by the superimposition of the frequency distribution of protein affinities reported in the literature and compiled in the kinetics database KOFFI [ 30 ].…”

Section: Resultsmentioning

confidence: 99%

“…Dots identify the datapoints corresponding to individual simulations. The right y -axis shows a histogram of the relative frequency of protein interactions in the KOFFI database with their affinities per order of magnitude [ 30 ]. The KOFFI database collects binding kinetics data from biomolecular interactions from the literature.…”

Section: Figurementioning

confidence: 99%

A Computational Investigation of In Vivo Cytosolic Protein Delivery for Cancer Therapy

et al. 2021

View full text Add to dashboard Cite

The ability to specifically block or degrade cytosolic targets using therapeutic proteins would bring tremendous therapeutic opportunities in cancer therapy. Over the last few years, significant progress has been made with respect to tissue targeting, cytosolic delivery, and catalytic inactivation of targets, placing this aim within reach. Here, we developed a mathematical model specifically built for the evaluation of approaches towards cytosolic protein delivery, involving all steps from systemic administration to translocation into the cytosol and target engagement. Focusing on solid cancer tissues, we utilized the model to investigate the effects of microvascular permeability, receptor affinity, the cellular density of targeted receptors, as well as the mode of activity (blocking/degradation) on therapeutic potential. Our analyses provide guidance for the rational optimization of protein design for enhanced activity and highlight the importance of tuning the receptor affinity as a function of receptor density as well as the receptor internalization rate. Furthermore, we provide quantitative insights into how enzymatic cargoes can enhance the distribution, extent, and duration of therapeutic activity, already at very low catalytic rates. Our results illustrate that with current protein engineering approaches, the goal of delivery of cytosolic delivery of proteins for therapeutic effects is well within reach.

show abstract

“…Also, the establishment of a baseline is essential for prediction tasks to improve results and make the validation process consistent and comparable. Existing k off databases such as SKEMPI 55 and KOFFI-DB 29 are mainly designed for protein−protein interactions, while BindingDB and KDBI 56 contain partial data on kinetic rate constants. 29 For residence time prediction, no prediction tools nor a special database with protein−small molecule binding kinetics and their 3D structures has been created; thus, comparative studies have not been performed yet.…”

Section: Resultsmentioning

confidence: 99%

“…ML methods can perform orders of magnitude faster than conventional cheminformatics computations and are also capable of automatically finding and estimating hidden relationships. , At the same time, ML requires obtaining a justified size of the training dataset in order to be useful and perform well. , Currently, binding kinetics databases are challenging to compile because residence time studies are uncommon and more experimental reports are yet to come. Also, the lack of standardized publication guidelines for kinetics experiment results renders it inefficient to extract them automatically . In one notable attempt to build such a database, the authors utilized a method for collaborative manual curation using a custom web-based annotation tool .…”

Section: Introductionmentioning

confidence: 99%

Baseline Model for Predicting Protein–Ligand Unbinding Kinetics through Machine Learning

Amangeldiuly

Karlov

Fedorov

2020

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Derivation of structure–kinetics relationships can help rational design and development of new small-molecule drug candidates with desired residence times. Efforts are now being directed toward the development of efficient computational methods. Currently, there is a lack of solid, high-throughput binding kinetics prediction approaches on bigger datasets. We present a prediction method for binding kinetics based on the machine learning analysis of protein–ligand structural features, which can serve as a baseline for more sophisticated methods utilizing molecular dynamics (MD). We showed that the random forest algorithm is capable of learning the protein binding site secondary structure and backbone/side-chain features to predict the binding kinetics of protein–ligand complexes but still with inferior performance to that of MD-based descriptor analysis. MD simulations had been applied to a limited number of targets and a series of ligands in terms of kinetics analysis, and we believe that the developed approach may guide new studies. The method was trained on a newly curated database of 501 protein–ligand unbinding rate constants, which can also be used for testing and training the binding kinetics prediction models.

show abstract

KOFFI and Anabel 2.0—a new binding kinetics database and its integration in an open-source binding analysis software

Cited by 10 publications

References 19 publications

Public Data Set of Protein–Ligand Dissociation Kinetic Constants for Quantitative Structure–Kinetics Relationship Studies

Public Data Set of Protein–Ligand Dissociation Kinetic Constants for Quantitative Structure–Kinetics Relationship Studies

A Computational Investigation of In Vivo Cytosolic Protein Delivery for Cancer Therapy

Baseline Model for Predicting Protein–Ligand Unbinding Kinetics through Machine Learning

Contact Info

Product

Resources

About