John J. Kozak
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
John J. Kozak has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
Harry B. Gray
Beckman Institute, California Institute of Technology, Pasadena, CA, 91125, USA
5 shared publicationsJay R. Winkler
Beckman Institute, California Institute of Technology, Pasadena, USA
5 shared publications- Enrique Abad
Departamento de Física Aplicada and Instituto de Computación Científica Avanzada (ICCAEX), Centro Universitario de Mérida, Universidad de Extremadura, E-06800 Mérida, Spain
4 shared publications - Harry B. Gray
California Institute of Technology
4 shared publications - Tim Abil
College of Computing and Digital Media DePaul University, 243 South Wabash, Chicago, Illinois 60604-2301, USA
4 shared publications G. Nicolis
Interdisciplinary Center for Nonlinear Phenomena and Complex Systems , Université Libre de Bruxelles , Campus Plaine, CP 231, bd du Triomphe, 1050 Brussels , Belgium
3 shared publications
Latest Publications
- 24 November 2020
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 124 (49) , 11067-11071
Conjecture on the Design of Helical Proteins
In an important advance in our understanding of protein folding, Wolynes and Onuchic found that the frustration ratio, Tf/Ts, for funneled energy landscapes is Tf/Ts ∼1.6. In our recent work on four heme proteins, we showed that when a protein unfolds from the native state to an early unfolded state, the degree of departure is characterized by a ratio f ∼1.6, where f is a measure of the elongation of n-residue segments of the polypeptide chain. Our analysis, which accounts for this apparent similarity in calculated signatures, is based on a logistic-map model of unfolding. We offer an important take home for the de novo protein synthesis community: in order to increase the probability of obtaining good quality crystals, nearest-neighbor repulsive interactions between adjacent residues (or sequences of residues) in the polypeptide chain must be propagated correctly.
- 24 November 2020
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 124 (49) , 11067-11071
Conjecture on the Design of Helical Proteins
In an important advance in our understanding of protein folding, Wolynes and Onuchic found that the frustration ratio, Tf/Ts, for funneled energy landscapes is Tf/Ts ∼1.6. In our recent work on four heme proteins, we showed that when a protein unfolds from the native state to an early unfolded state, the degree of departure is characterized by a ratio f ∼1.6, where f is a measure of the elongation of n-residue segments of the polypeptide chain. Our analysis, which accounts for this apparent similarity in calculated signatures, is based on a logistic-map model of unfolding. We offer an important take home for the de novo protein synthesis community: in order to increase the probability of obtaining good quality crystals, nearest-neighbor repulsive interactions between adjacent residues (or sequences of residues) in the polypeptide chain must be propagated correctly.
- 10 June 2019
- Published by American Physical Society (APS) in Physical Review E
- Vol. 99 (6) , 062110
First-passage properties of mortal random walks: Ballistic behavior, effective reduction of dimensionality, and scaling functions for hierarchical graphs
We consider a mortal random walker on a family of hierarchical graphs in the presence of some trap sites. The configuration comprising the graph, the starting point of the walk, and the locations of the trap sites is taken to be exactly self-similar as one goes from one generation of the family to the next. Under these circumstances, the total probability that the walker hits a trap is determined exactly as a function of the single-step survival probability of the mortal walker. On the generation graph of the family, this probability is shown to be given by the iterate of a certain scaling function or map . The properties of the map then determine, in each case, the behavior of the trapping probability, the mean time to trapping, the temporal scaling factor governing the random walk dimension on the graph, and other related properties. The formalism is illustrated for the cases of a linear hierarchical lattice and the Sierpinski graphs in two and three Euclidean dimensions. We find an effective reduction of the random walk dimensionality due to the ballistic behavior of the surviving particles induced by the mortality constraint. The relevance of this finding for experiments involving travel times of particles in diffusion-decay systems is discussed.- 10 June 2019
- Published by American Physical Society (APS) in Physical Review E
- Vol. 99 (6) , 062110
First-passage properties of mortal random walks: Ballistic behavior, effective reduction of dimensionality, and scaling functions for hierarchical graphs
We consider a mortal random walker on a family of hierarchical graphs in the presence of some trap sites. The configuration comprising the graph, the starting point of the walk, and the locations of the trap sites is taken to be exactly self-similar as one goes from one generation of the family to the next. Under these circumstances, the total probability that the walker hits a trap is determined exactly as a function of the single-step survival probability of the mortal walker. On the generation graph of the family, this probability is shown to be given by the iterate of a certain scaling function or map . The properties of the map then determine, in each case, the behavior of the trapping probability, the mean time to trapping, the temporal scaling factor governing the random walk dimension on the graph, and other related properties. The formalism is illustrated for the cases of a linear hierarchical lattice and the Sierpinski graphs in two and three Euclidean dimensions. We find an effective reduction of the random walk dimensionality due to the ballistic behavior of the surviving particles induced by the mortality constraint. The relevance of this finding for experiments involving travel times of particles in diffusion-decay systems is discussed.- Open Access
First-passage properties of mortal random walks: ballistic behavior, effective reduction of dimensionality, and scaling functions for hierarchical graphs
We consider a mortal random walker on a family of hierarchical graphs in the presence of some trap sites. The configuration comprising the graph, the starting point of the walk, and the locations of the trap sites is taken to be exactly self-similar as one goes from one generation of the family to the next. Under these circumstances, the total probability that the walker hits a trap is determined exactly as a function of the single-step survival probability $q$ of the mortal walker. On the $n^{\rm th}$ generation graph of the family, this probability is shown to be given by the $n^{\rm th}$ iterate of a certain scaling function or map $q \rightarrow f(q)$. The properties of the map then determine, in each case, the behavior of the trapping probability, the mean time to trapping, the temporal scaling factor governing the random walk dimension on the graph, and other related properties. The formalism is illustrated for the cases of a linear hierarchical lattice and the Sierpinski graphs in $2$ and $3$ Euclidean dimensions. We find an effective reduction of the random walk dimensionality due to the ballistic behavior of the surviving particles induced by the mortality constraint. The relevance of this finding for experiments involving travel times of particles in diffusion-decay systems is discussed.Open AccessFirst-passage properties of mortal random walks: ballistic behavior, effective reduction of dimensionality, and scaling functions for hierarchical graphs
We consider a mortal random walker on a family of hierarchical graphs in the presence of some trap sites. The configuration comprising the graph, the starting point of the walk, and the locations of the trap sites is taken to be exactly self-similar as one goes from one generation of the family to the next. Under these circumstances, the total probability that the walker hits a trap is determined exactly as a function of the single-step survival probability $q$ of the mortal walker. On the $n^{\rm th}$ generation graph of the family, this probability is shown to be given by the $n^{\rm th}$ iterate of a certain scaling function or map $q \rightarrow f(q)$. The properties of the map then determine, in each case, the behavior of the trapping probability, the mean time to trapping, the temporal scaling factor governing the random walk dimension on the graph, and other related properties. The formalism is illustrated for the cases of a linear hierarchical lattice and the Sierpinski graphs in $2$ and $3$ Euclidean dimensions. We find an effective reduction of the random walk dimensionality due to the ballistic behavior of the surviving particles induced by the mortality constraint. The relevance of this finding for experiments involving travel times of particles in diffusion-decay systems is discussed. - 24 August 2018
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 122 (49) , 11289-11294
Geometrical Description of Protein Structural Motifs
We present a geometrical method that can identify secondary structural motifs in proteins via angular correlations. The method uses crystal structure coordinates to calculate angular and radial signatures of each residue relative to an external reference point as the number of nearest-neighbor residues increases. We apply our approach to the blue copper protein amicyanin using the copper cofactor as the external reference point. We define a signature termed Δβ which describes the change in angular correlation as the span of nearest neighbor residues increases. We find that three turn regions of amicyanin harbor residues with Δβ near zero, while residues in other secondary structures have Δβ greater than zero: for β-strands, Δβ changes gradually residue by residue along the strand. Extension of our analysis to other blue copper proteins demonstrated that the noted structural trends are general. Importantly, our geometrical description of the folded protein accounts for all forces holding the structure together. Through this analysis, we identified some of the turns in amicyanin as symmetrical anchor points.
- 24 August 2018
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 122 (49) , 11289-11294
Geometrical Description of Protein Structural Motifs
We present a geometrical method that can identify secondary structural motifs in proteins via angular correlations. The method uses crystal structure coordinates to calculate angular and radial signatures of each residue relative to an external reference point as the number of nearest-neighbor residues increases. We apply our approach to the blue copper protein amicyanin using the copper cofactor as the external reference point. We define a signature termed Δβ which describes the change in angular correlation as the span of nearest neighbor residues increases. We find that three turn regions of amicyanin harbor residues with Δβ near zero, while residues in other secondary structures have Δβ greater than zero: for β-strands, Δβ changes gradually residue by residue along the strand. Extension of our analysis to other blue copper proteins demonstrated that the noted structural trends are general. Importantly, our geometrical description of the folded protein accounts for all forces holding the structure together. Through this analysis, we identified some of the turns in amicyanin as symmetrical anchor points.
- Open Access
Random Walks on Lattices. Influence of Competing Reaction Centers on Diffusion-Controlled Processes
We study diffusion-reaction processes on periodic square planar lattices and simple cubic (sc) lattices. Considered first is a single diffusing reactant undergoing an irreversible reaction upon first encounter with a stationary co-reactant ["one-walker (1W) problem"]. We then generalize this scenario to allow for a competing reaction, i.e., instantaneous trapping of the diffusing reactant with probability $s$ at any vacant site before interacting with a (stationary) co-reactant at a target site. We determine the mean walklength of the diffusing reactant until irreversible reaction occurs. We use generating functions and the theory of finite Markov processes, as well as MC simulations. To investigate the dependence of walklength on lattice size we compute the first, finite size corrections to the Green function of the sc lattice, and provide a Pad\'e approximation for this quantity. Finally, we consider the case where both reactant and co-reactant undergo synchronous nearest-neighbor displacements ["two-walker (2W) problem"]. In this case, reactant and co-reactant can individually be trapped with probability $s$ at any vacant lattice site, or can undergo an irreversible reaction on first encounter at any site. When $s=0$ we find that, both for the 1W and the 2W problem, for lattices with (approximately) the same number of sites, the mean walklength is smaller (and hence the reaction efficiency greater) in $d=3$ than in $d=2$. Increasing $s$ tends to reduce differences in system dimensionality, and distinctions between the 1W problem and the 2W problem. Our model provides a good starting point to develop studies on the efficiency of apparently diverse diffusion-reaction processes, such as diffusion on a partially poisoned catalytic substrate or photosynthetic trapping of excitations.Open AccessRandom Walks on Lattices. Influence of Competing Reaction Centers on Diffusion-Controlled Processes
We study diffusion-reaction processes on periodic square planar lattices and simple cubic (sc) lattices. Considered first is a single diffusing reactant undergoing an irreversible reaction upon first encounter with a stationary co-reactant ["one-walker (1W) problem"]. We then generalize this scenario to allow for a competing reaction, i.e., instantaneous trapping of the diffusing reactant with probability $s$ at any vacant site before interacting with a (stationary) co-reactant at a target site. We determine the mean walklength of the diffusing reactant until irreversible reaction occurs. We use generating functions and the theory of finite Markov processes, as well as MC simulations. To investigate the dependence of walklength on lattice size we compute the first, finite size corrections to the Green function of the sc lattice, and provide a Pad\'e approximation for this quantity. Finally, we consider the case where both reactant and co-reactant undergo synchronous nearest-neighbor displacements ["two-walker (2W) problem"]. In this case, reactant and co-reactant can individually be trapped with probability $s$ at any vacant lattice site, or can undergo an irreversible reaction on first encounter at any site. When $s=0$ we find that, both for the 1W and the 2W problem, for lattices with (approximately) the same number of sites, the mean walklength is smaller (and hence the reaction efficiency greater) in $d=3$ than in $d=2$. Increasing $s$ tends to reduce differences in system dimensionality, and distinctions between the 1W problem and the 2W problem. Our model provides a good starting point to develop studies on the efficiency of apparently diverse diffusion-reaction processes, such as diffusion on a partially poisoned catalytic substrate or photosynthetic trapping of excitations. - 3 August 2018
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 511, 336-357
Random walks on lattices. Influence of competing reaction centers on diffusion-controlled processes
- 3 August 2018
- Published by Elsevier in Physica A: Statistical Mechanics and its Applications
- Vol. 511, 336-357
Random walks on lattices. Influence of competing reaction centers on diffusion-controlled processes
- 31 December 2013
- Published by Springer Nature in JBIC Journal of Biological Inorganic Chemistry
- Vol. 19 (4-5) , 555-563
A Euclidean perspective on the unfolding of azurin: chain motion
We present a new approach to visualizing and quantifying the displacement of segments of Pseudomonas aeruginosa azurin in the early stages of denaturation. Our method is based on a geometrical method developed previously by the authors, and elaborated extensively for azurin. In this study, we quantify directional changes in three α-helical regions, two regions having β-strand residues, and three unstructured regions of azurin. Snapshots of these changes as the protein unfolds are displayed and described quantitatively by introducing a scaling diagnostic. In accord with molecular dynamics simulations, we show that the long α-helix in azurin (residues 54–67) is displaced from the polypeptide scaffolding and then pivots first in one direction, and then in the opposite direction as the protein continues to unfold. The two β-strand chains remain essentially intact and, except in the earliest stages, move in tandem. We show that unstructured regions 72–81 and 84–91, hinged by β-strand residues 82–83, pivot oppositely. The region comprising residues 72–91 (40 % hydrophobic and 16 % of the 128 total residues) forms an effectively stationary region that persists as the protein unfolds. This static behavior is a consequence of a dynamic balance between the competing motion of two segments, residues 72–81 and 84–91.
- 31 December 2013
- Published by Springer Nature in JBIC Journal of Biological Inorganic Chemistry
- Vol. 19 (4-5) , 555-563
A Euclidean perspective on the unfolding of azurin: chain motion
We present a new approach to visualizing and quantifying the displacement of segments of Pseudomonas aeruginosa azurin in the early stages of denaturation. Our method is based on a geometrical method developed previously by the authors, and elaborated extensively for azurin. In this study, we quantify directional changes in three α-helical regions, two regions having β-strand residues, and three unstructured regions of azurin. Snapshots of these changes as the protein unfolds are displayed and described quantitatively by introducing a scaling diagnostic. In accord with molecular dynamics simulations, we show that the long α-helix in azurin (residues 54–67) is displaced from the polypeptide scaffolding and then pivots first in one direction, and then in the opposite direction as the protein continues to unfold. The two β-strand chains remain essentially intact and, except in the earliest stages, move in tandem. We show that unstructured regions 72–81 and 84–91, hinged by β-strand residues 82–83, pivot oppositely. The region comprising residues 72–91 (40 % hydrophobic and 16 % of the 128 total residues) forms an effectively stationary region that persists as the protein unfolds. This static behavior is a consequence of a dynamic balance between the competing motion of two segments, residues 72–81 and 84–91.
- 26 November 2013
- Published by American Physical Society (APS) in Physical Review E
- Vol. 88 (5) , 052139
Analytic expression for the mean time to absorption for a random walker on the Sierpinski fractal. III. The effect of non-nearest-neighbor jumps
We present exact, analytic results for the mean time to trapping of a random walker on the class of deterministic Sierpinski graphs embedded in $d\phantom{\rule{0.16em}{0ex}}$\ge${}2$ Euclidean dimensions, when both nearest-neighbor (NN) and next-nearest-neighbor (NNN) jumps are included. Mean first-passage times are shown to be modified significantly as a consequence of the fact that NNN transitions connect fractals of two consecutive generations.- 26 November 2013
- Published by American Physical Society (APS) in Physical Review E
- Vol. 88 (5) , 052139
Analytic expression for the mean time to absorption for a random walker on the Sierpinski fractal. III. The effect of non-nearest-neighbor jumps
We present exact, analytic results for the mean time to trapping of a random walker on the class of deterministic Sierpinski graphs embedded in $d\phantom{\rule{0.16em}{0ex}}$\ge${}2$ Euclidean dimensions, when both nearest-neighbor (NN) and next-nearest-neighbor (NNN) jumps are included. Mean first-passage times are shown to be modified significantly as a consequence of the fact that NNN transitions connect fractals of two consecutive generations.- 15 March 2013
- Published by Springer Nature
- p. 579-586
Asymptotic Scaling for Euclidean Lattices
- 23 January 2013
- Published by Taylor & Francis in Molecular Physics
- Vol. 111 (7) , 922-929
A Euclidean perspective on the unfolding of azurin: spatial correlations
We investigate the stability to structural perturbation of Pseudomonas aeruginosa azurin using a previously developed geometric model. Our analysis considers Ru(2,2′,6′,2″-terpyridine)(1,10-phenanthroline)(His83)-labelled wild-type azurin and five variants with mutations to Cu-ligating residues. We find that in the early stages of unfolding, the β-strands exhibit the most structural stability. The conserved residues comprising the hydrophobic core are dislocated only after nearly complete unfolding of the β-barrel. Attachment of the Ru-complex at His83 does not destabilize the protein fold, despite causing some degree of structural rearrangement. Replacing the Cys112 and/or Met121 Cu ligands does not affect the conformational integrity of the protein. Notably, these results are in accord with experimental evidence, as well as molecular dynamics simulations of the denaturation of azurin.
- 23 January 2013
- Published by Taylor & Francis in Molecular Physics
- Vol. 111 (7) , 922-929
A Euclidean perspective on the unfolding of azurin: spatial correlations
We investigate the stability to structural perturbation of Pseudomonas aeruginosa azurin using a previously developed geometric model. Our analysis considers Ru(2,2′,6′,2″-terpyridine)(1,10-phenanthroline)(His83)-labelled wild-type azurin and five variants with mutations to Cu-ligating residues. We find that in the early stages of unfolding, the β-strands exhibit the most structural stability. The conserved residues comprising the hydrophobic core are dislocated only after nearly complete unfolding of the β-barrel. Attachment of the Ru-complex at His83 does not destabilize the protein fold, despite causing some degree of structural rearrangement. Replacing the Cys112 and/or Met121 Cu ligands does not affect the conformational integrity of the protein. Notably, these results are in accord with experimental evidence, as well as molecular dynamics simulations of the denaturation of azurin.