Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Basic weighted ensemble protocol and system setup. (A) Illustration of the WE protocol for a membrane permeability simulation in which a one-dimensional progress coordinate z ( eq ) is divided into bins, and iterative rounds of dynamics propagation and a resampling procedure are performed with the goal of providing even coverage along the coordinate. As seen in the upper left, two trajectories ( solid dots ) originating from the left-most bin each occupy a previously empty bin after N rounds of dynamics ( curved arrows ). The resampling procedure then replicates the trajectories in these newly occupied bins to maintain a target number of two trajectories per bin. (B) Simulation workflow used by permeability floe to directly calculate permeability coefficients, including one round of WE resampling (using WESTPA) and dynamics propagation for each WE iteration (see Figure S1 for further details). (C) Snapshot of the simulation system from a trajectory of a “rule of five” permeate, sotalol, crossing the periodic membrane. All water molecules have been removed for clarity. L z is the z -component of the simulation box, while z ′ is the distance from the center of mass of the permeate and the center of mass of the membrane ( straight blue lines ).

Article Snippet: OpenEye Permeability Floe is available on the academic stack of Orion, which is also available upon request.

Techniques: Membrane, Permeability

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Estimated permeability coefficients of tacrine. (A) Chemical structure of tacrine. (B) Average (log) permeability coefficients (with 95% confidence intervals) calculated from each of the four WE simulations with different protocols [see the legend in panel (C)]. The gray dashed lines indicate the observed values from MDCK-LE and PAMPA experiments. Dashed arrows indicate the raised permeability estimates by applying WESS reweighting to Protocol 1 and 3 after the simulations were completed (50 ns). (C) Time evolution of estimated logarithm of the permeability coefficients (cm/s) for tacrine from four WE simulations, using a fixed binning scheme with (regular, blue ) or without WESS ( orange ), and the MAB scheme with ( green ) or without WESS ( purple ), respectively. The solid lines indicate the mean values of the estimates, and the shaded areas indicate 95% confidence intervals. The dashed gray line indicates the permeability coefficient measured by an MDCK-LE experiment. The molecular time is represented as N τ, where N is the number of WE iterations and τ is the fixed time interval (100 ps) of each WE iteration. (D) Snapshots of the tacrine molecule ( cyan ) passing through the lipid bilayer ( gray ) at selected molecular times. Water molecules in close contact with the molecule are highlighted in magenta .

Article Snippet: OpenEye Permeability Floe is available on the academic stack of Orion, which is also available upon request.

Techniques: Permeability, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Membrane permeabilities (Log P m ) calculated using various WE simulation protocols for tacrine, sotalol, and zacopride. Uncertainties represent standard deviations, which are evaluated as 1/4th the difference between the 95% CI upper bound and the lower bound. Experimentally measured values are shown in gray [MDCK-LE: Dickson et al. (2019)], PAMPA for tacrine: Katt et al. (2016), and PAMPA for sotalol: Liu et al. (2012). See also Table S1 in the Supporting Information.

Article Snippet: OpenEye Permeability Floe is available on the academic stack of Orion, which is also available upon request.

Techniques: Membrane, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Efficiencies of Different WE Protocols in Predicting the Membrane Permeability (Log P m ) of Tacrine

Article Snippet: OpenEye Permeability Floe is available on the academic stack of Orion, which is also available upon request.

Techniques: Membrane, Permeability

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Basic weighted ensemble protocol and system setup. (A) Illustration of the WE protocol for a membrane permeability simulation in which a one-dimensional progress coordinate z ( eq ) is divided into bins, and iterative rounds of dynamics propagation and a resampling procedure are performed with the goal of providing even coverage along the coordinate. As seen in the upper left, two trajectories ( solid dots ) originating from the left-most bin each occupy a previously empty bin after N rounds of dynamics ( curved arrows ). The resampling procedure then replicates the trajectories in these newly occupied bins to maintain a target number of two trajectories per bin. (B) Simulation workflow used by permeability floe to directly calculate permeability coefficients, including one round of WE resampling (using WESTPA) and dynamics propagation for each WE iteration (see Figure S1 for further details). (C) Snapshot of the simulation system from a trajectory of a “rule of five” permeate, sotalol, crossing the periodic membrane. All water molecules have been removed for clarity. L z is the z -component of the simulation box, while z ′ is the distance from the center of mass of the permeate and the center of mass of the membrane ( straight blue lines ).

Article Snippet: Here, we present the results from fully automated permeability simulations performed on three “rule of five” molecules (tacrine, zacopride, and sotalol) using the OpenEye Permeability Floe package in the Orion cloud computing environment.

Techniques: Membrane, Permeability

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Estimated permeability coefficients of tacrine. (A) Chemical structure of tacrine. (B) Average (log) permeability coefficients (with 95% confidence intervals) calculated from each of the four WE simulations with different protocols [see the legend in panel (C)]. The gray dashed lines indicate the observed values from MDCK-LE and PAMPA experiments. Dashed arrows indicate the raised permeability estimates by applying WESS reweighting to Protocol 1 and 3 after the simulations were completed (50 ns). (C) Time evolution of estimated logarithm of the permeability coefficients (cm/s) for tacrine from four WE simulations, using a fixed binning scheme with (regular, blue ) or without WESS ( orange ), and the MAB scheme with ( green ) or without WESS ( purple ), respectively. The solid lines indicate the mean values of the estimates, and the shaded areas indicate 95% confidence intervals. The dashed gray line indicates the permeability coefficient measured by an MDCK-LE experiment. The molecular time is represented as N τ, where N is the number of WE iterations and τ is the fixed time interval (100 ps) of each WE iteration. (D) Snapshots of the tacrine molecule ( cyan ) passing through the lipid bilayer ( gray ) at selected molecular times. Water molecules in close contact with the molecule are highlighted in magenta .

Article Snippet: Here, we present the results from fully automated permeability simulations performed on three “rule of five” molecules (tacrine, zacopride, and sotalol) using the OpenEye Permeability Floe package in the Orion cloud computing environment.

Techniques: Permeability, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Membrane permeabilities (Log P m ) calculated using various WE simulation protocols for tacrine, sotalol, and zacopride. Uncertainties represent standard deviations, which are evaluated as 1/4th the difference between the 95% CI upper bound and the lower bound. Experimentally measured values are shown in gray [MDCK-LE: Dickson et al. (2019)], PAMPA for tacrine: Katt et al. (2016), and PAMPA for sotalol: Liu et al. (2012). See also Table S1 in the Supporting Information.

Article Snippet: Here, we present the results from fully automated permeability simulations performed on three “rule of five” molecules (tacrine, zacopride, and sotalol) using the OpenEye Permeability Floe package in the Orion cloud computing environment.

Techniques: Membrane, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Efficiencies of Different WE Protocols in Predicting the Membrane Permeability (Log P m ) of Tacrine

Article Snippet: Here, we present the results from fully automated permeability simulations performed on three “rule of five” molecules (tacrine, zacopride, and sotalol) using the OpenEye Permeability Floe package in the Orion cloud computing environment.

Techniques: Membrane, Permeability

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Basic weighted ensemble protocol and system setup. (A) Illustration of the WE protocol for a membrane permeability simulation in which a one-dimensional progress coordinate z ( eq ) is divided into bins, and iterative rounds of dynamics propagation and a resampling procedure are performed with the goal of providing even coverage along the coordinate. As seen in the upper left, two trajectories ( solid dots ) originating from the left-most bin each occupy a previously empty bin after N rounds of dynamics ( curved arrows ). The resampling procedure then replicates the trajectories in these newly occupied bins to maintain a target number of two trajectories per bin. (B) Simulation workflow used by permeability floe to directly calculate permeability coefficients, including one round of WE resampling (using WESTPA) and dynamics propagation for each WE iteration (see Figure S1 for further details). (C) Snapshot of the simulation system from a trajectory of a “rule of five” permeate, sotalol, crossing the periodic membrane. All water molecules have been removed for clarity. L z is the z -component of the simulation box, while z ′ is the distance from the center of mass of the permeate and the center of mass of the membrane ( straight blue lines ).

Article Snippet: The OpenEye Permeability floe in Orion contains a series of Cubes that each performs one of the following functions: system preparation, MD equilibration, WE simulation, and permeability analysis of the membrane-permeate system (see Figure B for an example of the flow relationship diagram of the compute kernels and above sections for details).

Techniques: Membrane, Permeability

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Estimated permeability coefficients of tacrine. (A) Chemical structure of tacrine. (B) Average (log) permeability coefficients (with 95% confidence intervals) calculated from each of the four WE simulations with different protocols [see the legend in panel (C)]. The gray dashed lines indicate the observed values from MDCK-LE and PAMPA experiments. Dashed arrows indicate the raised permeability estimates by applying WESS reweighting to Protocol 1 and 3 after the simulations were completed (50 ns). (C) Time evolution of estimated logarithm of the permeability coefficients (cm/s) for tacrine from four WE simulations, using a fixed binning scheme with (regular, blue ) or without WESS ( orange ), and the MAB scheme with ( green ) or without WESS ( purple ), respectively. The solid lines indicate the mean values of the estimates, and the shaded areas indicate 95% confidence intervals. The dashed gray line indicates the permeability coefficient measured by an MDCK-LE experiment. The molecular time is represented as N τ, where N is the number of WE iterations and τ is the fixed time interval (100 ps) of each WE iteration. (D) Snapshots of the tacrine molecule ( cyan ) passing through the lipid bilayer ( gray ) at selected molecular times. Water molecules in close contact with the molecule are highlighted in magenta .

Article Snippet: The OpenEye Permeability floe in Orion contains a series of Cubes that each performs one of the following functions: system preparation, MD equilibration, WE simulation, and permeability analysis of the membrane-permeate system (see Figure B for an example of the flow relationship diagram of the compute kernels and above sections for details).

Techniques: Permeability, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Membrane permeabilities (Log P m ) calculated using various WE simulation protocols for tacrine, sotalol, and zacopride. Uncertainties represent standard deviations, which are evaluated as 1/4th the difference between the 95% CI upper bound and the lower bound. Experimentally measured values are shown in gray [MDCK-LE: Dickson et al. (2019)], PAMPA for tacrine: Katt et al. (2016), and PAMPA for sotalol: Liu et al. (2012). See also Table S1 in the Supporting Information.

Article Snippet: The OpenEye Permeability floe in Orion contains a series of Cubes that each performs one of the following functions: system preparation, MD equilibration, WE simulation, and permeability analysis of the membrane-permeate system (see Figure B for an example of the flow relationship diagram of the compute kernels and above sections for details).

Techniques: Membrane, PAMPA Assay

Journal: Journal of Chemical Information and Modeling

Article Title: Mechanistic Insights into Passive Membrane Permeability of Drug-like Molecules from a Weighted Ensemble of Trajectories

doi: 10.1021/acs.jcim.1c01540

Figure Lengend Snippet: Efficiencies of Different WE Protocols in Predicting the Membrane Permeability (Log P m ) of Tacrine

Article Snippet: The OpenEye Permeability floe in Orion contains a series of Cubes that each performs one of the following functions: system preparation, MD equilibration, WE simulation, and permeability analysis of the membrane-permeate system (see Figure B for an example of the flow relationship diagram of the compute kernels and above sections for details).

Techniques: Membrane, Permeability