Machine Learning-Driven Methods for Nanobody Affinity Prediction

Because of their high affinity, specificity, and environmental stability, nanobodies (Nbs) have continuously received attention from the field of biological research. However, it is tough work to obtain high-affinity Nbs using experimental methods. In the current study, 12 machine learning algorithm...

Full description

Saved in:
Bibliographic Details
Main Author: Hua Feng (234718) (author)
Other Authors: Xuefeng Sun (408801) (author), Ning Li (45258) (author), Qian Xu (119888) (author), Qin Li (36669) (author), Shenli Zhang (3236712) (author), Guangxu Xing (336317) (author), Gaiping Zhang (106747) (author), Fangyu Wang (4279168) (author)
Published: 2024
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1852025113734545408
author Hua Feng (234718)
author2 Xuefeng Sun (408801)
Ning Li (45258)
Qian Xu (119888)
Qin Li (36669)
Shenli Zhang (3236712)
Guangxu Xing (336317)
Gaiping Zhang (106747)
Fangyu Wang (4279168)
author2_role author
author
author
author
author
author
author
author
author_facet Hua Feng (234718)
Xuefeng Sun (408801)
Ning Li (45258)
Qian Xu (119888)
Qin Li (36669)
Shenli Zhang (3236712)
Guangxu Xing (336317)
Gaiping Zhang (106747)
Fangyu Wang (4279168)
author_role author
dc.creator.none.fl_str_mv Hua Feng (234718)
Xuefeng Sun (408801)
Ning Li (45258)
Qian Xu (119888)
Qin Li (36669)
Shenli Zhang (3236712)
Guangxu Xing (336317)
Gaiping Zhang (106747)
Fangyu Wang (4279168)
dc.date.none.fl_str_mv 2024-11-19T17:07:08Z
dc.identifier.none.fl_str_mv 10.1021/acsomega.4c09718.s001
dc.relation.none.fl_str_mv https://figshare.com/articles/dataset/Machine_Learning-Driven_Methods_for_Nanobody_Affinity_Prediction/27855283
dc.rights.none.fl_str_mv CC BY-NC 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biophysics
Biochemistry
Molecular Biology
Ecology
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
continuously received attention
eight noncovalent interactions
true affinitive nbs
showed higher sensitivity
predicting nonaffinitive nbs
mcc values ).
good model robustness
future nonaffinitive nbs
two stacked models
key noncovalent interactions
nanobody affinity prediction
effectively predict whether
current study provides
four optimized models
effectively predict
current study
optimized models
models showed
associated interactions
model comparison
key factors
5 ).
tough work
potential patterns
obtain high
nine uncorrelated
nb drugs
machine learning
intended ligands
hydrogen bonding
high affinity
first time
f1 score
environmental stability
driven methods
design process
could improve
correlation coefficient
biological research
dc.title.none.fl_str_mv Machine Learning-Driven Methods for Nanobody Affinity Prediction
dc.type.none.fl_str_mv Dataset
info:eu-repo/semantics/publishedVersion
dataset
description Because of their high affinity, specificity, and environmental stability, nanobodies (Nbs) have continuously received attention from the field of biological research. However, it is tough work to obtain high-affinity Nbs using experimental methods. In the current study, 12 machine learning algorithms were compared in parallel to explore the potential patterns between Nb–ligand affinity and eight noncovalent interactions. After model comparison and optimization, four optimized models (SVMrB, RotFB, RFB, and C50B) and two stacked models (StackKNN and StackRF) based on nine uncorrelated (correlation coefficient <0.65) optimized models were selected. All the models showed an accuracy of around 0.70 and high specificity. Compared to the other models, RotFB and RFB were not capable of predicting nonaffinitive Nbs with lower precision (<0.44) but showed higher sensitivity at 0.6761 and 0.3521 and good model robustness (F1 score and MCC values). On the contrary, SVMrB, C50B, and StackKNN were able to effectively predict the future nonaffinitive Nbs (specificity >0.92) and reduce the number of true affinitive Nbs (precision >0.5). On the other hand, StackRF showed intermediate model performance. Furthermore, an in-depth feature analysis indicated that hydrogen bonding and aromatic-associated interactions were the key noncovalent interactions in determining Nb–ligand binding affinity. In summary, the current study provides, for the first time, a tool that can effectively predict whether there is an affinity between nanobodies and their intended ligands and explores the key factors that influence their affinity, which could improve the screening and design process of Nbs and accelerate the development of Nb drugs and applications.
eu_rights_str_mv openAccess
id Manara_eaee6caa1cdb2db930a1465451fa72b5
identifier_str_mv 10.1021/acsomega.4c09718.s001
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/27855283
publishDate 2024
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY-NC 4.0
spelling Machine Learning-Driven Methods for Nanobody Affinity PredictionHua Feng (234718)Xuefeng Sun (408801)Ning Li (45258)Qian Xu (119888)Qin Li (36669)Shenli Zhang (3236712)Guangxu Xing (336317)Gaiping Zhang (106747)Fangyu Wang (4279168)BiophysicsBiochemistryMolecular BiologyEcologyBiological Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedcontinuously received attentioneight noncovalent interactionstrue affinitive nbsshowed higher sensitivitypredicting nonaffinitive nbsmcc values ).good model robustnessfuture nonaffinitive nbstwo stacked modelskey noncovalent interactionsnanobody affinity predictioneffectively predict whethercurrent study providesfour optimized modelseffectively predictcurrent studyoptimized modelsmodels showedassociated interactionsmodel comparisonkey factors5 ).tough workpotential patternsobtain highnine uncorrelatednb drugsmachine learningintended ligandshydrogen bondinghigh affinityfirst timef1 scoreenvironmental stabilitydriven methodsdesign processcould improvecorrelation coefficientbiological researchBecause of their high affinity, specificity, and environmental stability, nanobodies (Nbs) have continuously received attention from the field of biological research. However, it is tough work to obtain high-affinity Nbs using experimental methods. In the current study, 12 machine learning algorithms were compared in parallel to explore the potential patterns between Nb–ligand affinity and eight noncovalent interactions. After model comparison and optimization, four optimized models (SVMrB, RotFB, RFB, and C50B) and two stacked models (StackKNN and StackRF) based on nine uncorrelated (correlation coefficient <0.65) optimized models were selected. All the models showed an accuracy of around 0.70 and high specificity. Compared to the other models, RotFB and RFB were not capable of predicting nonaffinitive Nbs with lower precision (<0.44) but showed higher sensitivity at 0.6761 and 0.3521 and good model robustness (F1 score and MCC values). On the contrary, SVMrB, C50B, and StackKNN were able to effectively predict the future nonaffinitive Nbs (specificity >0.92) and reduce the number of true affinitive Nbs (precision >0.5). On the other hand, StackRF showed intermediate model performance. Furthermore, an in-depth feature analysis indicated that hydrogen bonding and aromatic-associated interactions were the key noncovalent interactions in determining Nb–ligand binding affinity. In summary, the current study provides, for the first time, a tool that can effectively predict whether there is an affinity between nanobodies and their intended ligands and explores the key factors that influence their affinity, which could improve the screening and design process of Nbs and accelerate the development of Nb drugs and applications.2024-11-19T17:07:08ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1021/acsomega.4c09718.s001https://figshare.com/articles/dataset/Machine_Learning-Driven_Methods_for_Nanobody_Affinity_Prediction/27855283CC BY-NC 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/278552832024-11-19T17:07:08Z
spellingShingle Machine Learning-Driven Methods for Nanobody Affinity Prediction
Hua Feng (234718)
Biophysics
Biochemistry
Molecular Biology
Ecology
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
continuously received attention
eight noncovalent interactions
true affinitive nbs
showed higher sensitivity
predicting nonaffinitive nbs
mcc values ).
good model robustness
future nonaffinitive nbs
two stacked models
key noncovalent interactions
nanobody affinity prediction
effectively predict whether
current study provides
four optimized models
effectively predict
current study
optimized models
models showed
associated interactions
model comparison
key factors
5 ).
tough work
potential patterns
obtain high
nine uncorrelated
nb drugs
machine learning
intended ligands
hydrogen bonding
high affinity
first time
f1 score
environmental stability
driven methods
design process
could improve
correlation coefficient
biological research
status_str publishedVersion
title Machine Learning-Driven Methods for Nanobody Affinity Prediction
title_full Machine Learning-Driven Methods for Nanobody Affinity Prediction
title_fullStr Machine Learning-Driven Methods for Nanobody Affinity Prediction
title_full_unstemmed Machine Learning-Driven Methods for Nanobody Affinity Prediction
title_short Machine Learning-Driven Methods for Nanobody Affinity Prediction
title_sort Machine Learning-Driven Methods for Nanobody Affinity Prediction
topic Biophysics
Biochemistry
Molecular Biology
Ecology
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
continuously received attention
eight noncovalent interactions
true affinitive nbs
showed higher sensitivity
predicting nonaffinitive nbs
mcc values ).
good model robustness
future nonaffinitive nbs
two stacked models
key noncovalent interactions
nanobody affinity prediction
effectively predict whether
current study provides
four optimized models
effectively predict
current study
optimized models
models showed
associated interactions
model comparison
key factors
5 ).
tough work
potential patterns
obtain high
nine uncorrelated
nb drugs
machine learning
intended ligands
hydrogen bonding
high affinity
first time
f1 score
environmental stability
driven methods
design process
could improve
correlation coefficient
biological research