A deep encoder-decoder framework for identifying distinct ligand binding pathways

The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high di...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of chemical physics Jg. 158; H. 19
Hauptverfasser: Bandyopadhyay, Satyabrata, Mondal, Jagannath
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States 21.05.2023
Schlagworte:
ISSN:1089-7690, 1089-7690
Online-Zugang:Weitere Angaben
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high dimensional nature of the MD-derived trajectories often remains a barrier in the statistical elucidation of distinct ligand binding pathways due to the stochasticity inherent in the ligand's fluctuation in the solution and around the receptor. Here, we demonstrate that an autoencoder based deep neural network, trained using an objective input feature of a large matrix of residue-ligand distances, can efficiently produce an optimal low-dimensional latent space that stores necessary information on the ligand-binding event. In particular, for a system of L99A mutant of T4 lysozyme interacting with its native ligand, benzene, this deep encoder-decoder framework automatically identifies multiple distinct recognition pathways, without requiring user intervention. The intermediates involve the spatially discrete location of the ligand in different helices of the protein before its eventual recognition of native pose. The compressed subspace derived from the autoencoder provides a quantitatively accurate measure of the free energy and kinetics of ligand binding to the native pocket. The investigation also recommends that while a linear dimensional reduction technique, such as time-structured independent component analysis, can do a decent job of state-space decomposition in cases where the intermediates are long-lived, autoencoder is the method of choice in systems where transient, low-populated intermediates can lead to multiple ligand-binding pathways.
AbstractList The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high dimensional nature of the MD-derived trajectories often remains a barrier in the statistical elucidation of distinct ligand binding pathways due to the stochasticity inherent in the ligand's fluctuation in the solution and around the receptor. Here, we demonstrate that an autoencoder based deep neural network, trained using an objective input feature of a large matrix of residue-ligand distances, can efficiently produce an optimal low-dimensional latent space that stores necessary information on the ligand-binding event. In particular, for a system of L99A mutant of T4 lysozyme interacting with its native ligand, benzene, this deep encoder-decoder framework automatically identifies multiple distinct recognition pathways, without requiring user intervention. The intermediates involve the spatially discrete location of the ligand in different helices of the protein before its eventual recognition of native pose. The compressed subspace derived from the autoencoder provides a quantitatively accurate measure of the free energy and kinetics of ligand binding to the native pocket. The investigation also recommends that while a linear dimensional reduction technique, such as time-structured independent component analysis, can do a decent job of state-space decomposition in cases where the intermediates are long-lived, autoencoder is the method of choice in systems where transient, low-populated intermediates can lead to multiple ligand-binding pathways.
The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high dimensional nature of the MD-derived trajectories often remains a barrier in the statistical elucidation of distinct ligand binding pathways due to the stochasticity inherent in the ligand's fluctuation in the solution and around the receptor. Here, we demonstrate that an autoencoder based deep neural network, trained using an objective input feature of a large matrix of residue-ligand distances, can efficiently produce an optimal low-dimensional latent space that stores necessary information on the ligand-binding event. In particular, for a system of L99A mutant of T4 lysozyme interacting with its native ligand, benzene, this deep encoder-decoder framework automatically identifies multiple distinct recognition pathways, without requiring user intervention. The intermediates involve the spatially discrete location of the ligand in different helices of the protein before its eventual recognition of native pose. The compressed subspace derived from the autoencoder provides a quantitatively accurate measure of the free energy and kinetics of ligand binding to the native pocket. The investigation also recommends that while a linear dimensional reduction technique, such as time-structured independent component analysis, can do a decent job of state-space decomposition in cases where the intermediates are long-lived, autoencoder is the method of choice in systems where transient, low-populated intermediates can lead to multiple ligand-binding pathways.The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological activity. While Molecular Dynamics (MD) simulations have emerged as the method of choice for modeling protein-ligand binding events, the high dimensional nature of the MD-derived trajectories often remains a barrier in the statistical elucidation of distinct ligand binding pathways due to the stochasticity inherent in the ligand's fluctuation in the solution and around the receptor. Here, we demonstrate that an autoencoder based deep neural network, trained using an objective input feature of a large matrix of residue-ligand distances, can efficiently produce an optimal low-dimensional latent space that stores necessary information on the ligand-binding event. In particular, for a system of L99A mutant of T4 lysozyme interacting with its native ligand, benzene, this deep encoder-decoder framework automatically identifies multiple distinct recognition pathways, without requiring user intervention. The intermediates involve the spatially discrete location of the ligand in different helices of the protein before its eventual recognition of native pose. The compressed subspace derived from the autoencoder provides a quantitatively accurate measure of the free energy and kinetics of ligand binding to the native pocket. The investigation also recommends that while a linear dimensional reduction technique, such as time-structured independent component analysis, can do a decent job of state-space decomposition in cases where the intermediates are long-lived, autoencoder is the method of choice in systems where transient, low-populated intermediates can lead to multiple ligand-binding pathways.
Author Bandyopadhyay, Satyabrata
Mondal, Jagannath
Author_xml – sequence: 1
  givenname: Satyabrata
  orcidid: 0000-0002-0927-3326
  surname: Bandyopadhyay
  fullname: Bandyopadhyay, Satyabrata
  organization: Tata Institute of Fundamental Research, Center for Interdisciplinary Sciences, Hyderabad 500046, India
– sequence: 2
  givenname: Jagannath
  orcidid: 0000-0003-1090-5199
  surname: Mondal
  fullname: Mondal, Jagannath
  organization: Tata Institute of Fundamental Research, Center for Interdisciplinary Sciences, Hyderabad 500046, India
BackLink https://www.ncbi.nlm.nih.gov/pubmed/37184003$$D View this record in MEDLINE/PubMed
BookMark eNpNkEtLxDAUhYOMOA9d-AckSzcdb9ImTZbD4AsEEXRd0uRmjLZpbToM8-99jeDqOxw-zuLMySR2EQk5Z7BkIPMrsQRWCKbLIzJjoHRWSg2Tf3lK5im9AQAreXFCpnnJVAGQz8jTijrEnmK0ncMhc_hD6gfT4q4b3qnvBhocxjH4fYgb6kIaQ7QjbcLGREfrEN1335vxdWf26ZQce9MkPDtwQV5urp_Xd9nD4-39evWQ2RzUmBluBOalBi7qEqzn0jqQtkYjGedGSSskL5gC1FJ7q71wRglV1LrgBrThC3L5u9sP3ccW01i1IVlsGhOx26aKK5YrJSWXX-rFQd3WLbqqH0Jrhn31dwP_BMwsX2U
ContentType Journal Article
Copyright 2023 Author(s). Published under an exclusive license by AIP Publishing.
Copyright_xml – notice: 2023 Author(s). Published under an exclusive license by AIP Publishing.
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1063/5.0145197
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Chemistry
Physics
EISSN 1089-7690
ExternalDocumentID 37184003
Genre Journal Article
GroupedDBID ---
-DZ
-ET
-~X
123
1UP
2-P
29K
4.4
53G
5VS
85S
AAAAW
AABDS
AAEUA
AAPUP
AAYIH
ABJGX
ABPPZ
ABZEH
ACBRY
ACLYJ
ACNCT
ACZLF
ADCTM
ADMLS
AEJMO
AENEX
AFATG
AFHCQ
AGKCL
AGLKD
AGMXG
AGTJO
AHSDT
AJJCW
AJQPL
ALEPV
ALMA_UNASSIGNED_HOLDINGS
AQWKA
ATXIE
AWQPM
BDMKI
BPZLN
CGR
CS3
CUY
CVF
D-I
DU5
EBS
ECM
EIF
ESX
F5P
FDOHQ
FFFMQ
HAM
M6X
M71
M73
N9A
NPM
NPSNA
O-B
P2P
RIP
RNS
RQS
TN5
TWZ
UPT
WH7
YQT
YZZ
~02
7X8
AAGWI
ID FETCH-LOGICAL-c308t-a2a5e379025b70cf26cd06cbea6122a86c5624180e969fc9f5da8584b942a09a2
IEDL.DBID 7X8
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001017167500017&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1089-7690
IngestDate Thu Jul 10 21:44:14 EDT 2025
Wed Feb 19 02:23:11 EST 2025
IsPeerReviewed true
IsScholarly true
Issue 19
Language English
License 2023 Author(s). Published under an exclusive license by AIP Publishing.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c308t-a2a5e379025b70cf26cd06cbea6122a86c5624180e969fc9f5da8584b942a09a2
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-0927-3326
0000-0003-1090-5199
PMID 37184003
PQID 2813886626
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2813886626
pubmed_primary_37184003
PublicationCentury 2000
PublicationDate 2023-May-21
20230521
PublicationDateYYYYMMDD 2023-05-21
PublicationDate_xml – month: 05
  year: 2023
  text: 2023-May-21
  day: 21
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle The Journal of chemical physics
PublicationTitleAlternate J Chem Phys
PublicationYear 2023
SSID ssj0001724
Score 2.440442
Snippet The pathway(s) that a ligand would adopt en route to its trajectory to the native pocket of the receptor protein act as a key determinant of its biological...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
SubjectTerms Ligands
Molecular Dynamics Simulation
Neural Networks, Computer
Protein Binding
Proteins - chemistry
Title A deep encoder-decoder framework for identifying distinct ligand binding pathways
URI https://www.ncbi.nlm.nih.gov/pubmed/37184003
https://www.proquest.com/docview/2813886626
Volume 158
WOSCitedRecordID wos001017167500017&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3LSgMxFA1qFd34qK_6IoLb0EwyzWMlpVhcaKmg0N2QZBIpyLR2quLfm2Rm6EoQ3ExWA5NL5uYk595zALgxVkrbSwTKrSQoJdghQYxGTjpmg-R3Uom4PvDRSEwmclxfuJV1WWWTE2Oizmcm3JF3iUioEMzj79v5OwquUYFdrS001kGLeigTSrr4ZKUW7jfntCqwl4j7Y2CjLMRoN4h1BmUV_juyjDvMcO-_37YPdmtsCfvVYjgAa7Zog-1BY-nWBlux3tOUh-CpD3Nr5zDIWOZ24WMYR-iaWi3owSycxi7e2AkF85AMCrOEb9NXVeRQT2M_DAyWxl_quzwCL8O758E9qt0VkKFYLJEiqmcpDzSj5tg4wkyOmdFWedBDlGDGQ6M0EdhKJp2Rrpcr4eGKlilRWCpyDDaKWWFPAWRCcx34OBkBCdVSOEdTrITEzFreAddN3DI_40BJqMLOPspsFbkOOKmCn80rmY2M8nD6xPTsD2-fg53gAx9ofZJcgJbz_669BJvmczktF1dxWfjnaPz4A1BRwWk
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+deep+encoder-decoder+framework+for+identifying+distinct+ligand+binding+pathways&rft.jtitle=The+Journal+of+chemical+physics&rft.au=Bandyopadhyay%2C+Satyabrata&rft.au=Mondal%2C+Jagannath&rft.date=2023-05-21&rft.issn=1089-7690&rft.eissn=1089-7690&rft.volume=158&rft.issue=19&rft_id=info:doi/10.1063%2F5.0145197&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1089-7690&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1089-7690&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1089-7690&client=summon