Limits...
RaTrav: a tool for calculating mean first-passage times on biochemical networks.

Torchala M, Chelminiak P, Kurzynski M, Bates PA - BMC Syst Biol (2013)

Bottom Line: The RaTrav tool can then be applied in order to compute desired MFPTs.For the provided examples, we were able to find the favourable binding path within a protein-protein docking funnel and to calculate the degree of coupling for two chemical reactions catalysed simultaneously by the same protein enzyme.However, the list of possible applications is much wider.

View Article: PubMed Central - HTML - PubMed

Affiliation: Biomolecular Modelling Laboratory, Cancer Research UK London Research Institute, 44 Lincoln's Inn Fields, London WC2A 3LY, UK. Paul.Bates@cancer.org.uk.

ABSTRACT

Background: The concept of mean first-passage times (MFPTs) occupies an important place in the theory of stochastic processes, with the methods of their calculation being equally important in theoretical physics, chemistry and biology. We present here a software tool designed to support computational biology studies where Markovian dynamics takes place and MFPTs between initial and single or multiple final states in network-like systems are used. Two methods are made available for which their efficiency is strongly dependent on the topology of the defined network: the combinatorial Hill technique and the Monte Carlo simulation method.

Results: After a brief introduction to RaTrav, we highlight the utility of MFPT calculations by providing two examples (accompanied by Additional file 1) where they are deemed to be of importance: analysis of a protein-protein docking funnel and interpretation of the free energy transduction between two coupled enzymatic reactions controlled by the dynamics of transition between enzyme conformational states.

Conclusions: RaTrav is a versatile and easy to use software tool for calculating MFPTs across biochemical networks. The user simply prepares a text file with the structure of a given network, along with some additional basic parameters such as transition probabilities, waiting probabilities (if any) and local times (weights of edges), which define explicitly the stochastic dynamics on the network. The RaTrav tool can then be applied in order to compute desired MFPTs. For the provided examples, we were able to find the favourable binding path within a protein-protein docking funnel and to calculate the degree of coupling for two chemical reactions catalysed simultaneously by the same protein enzyme. However, the list of possible applications is much wider.

Show MeSH

Related in: MedlinePlus

The true positive binding funnel. The true positive binding funnel for the vitamin D-binding protein/actin complex (1KXP [22]) generated with the SwarmDock Server [20] by docking unbound receptor/ligand pair included in Benchmark 4.0 [25]. The protein-protein conformational states are numbered from ID 0 to 31. The letter indicates the quality of solution in accordance to CAPRI [24] classification as: M – medium quality (blue), A – acceptable quality (green), I – incorrect solution (red). Favourable paths are marked in red. Figure created with Gephi [26] based on Example1/1KXP.gml file (Additional file 1). See text for more details.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4222613&req=5

Figure 2: The true positive binding funnel. The true positive binding funnel for the vitamin D-binding protein/actin complex (1KXP [22]) generated with the SwarmDock Server [20] by docking unbound receptor/ligand pair included in Benchmark 4.0 [25]. The protein-protein conformational states are numbered from ID 0 to 31. The letter indicates the quality of solution in accordance to CAPRI [24] classification as: M – medium quality (blue), A – acceptable quality (green), I – incorrect solution (red). Favourable paths are marked in red. Figure created with Gephi [26] based on Example1/1KXP.gml file (Additional file 1). See text for more details.

Mentions: The initial network of 32 conformational states, generated by the SwarmDock server, is depicted in Figure 2. The assigned quality of each state, that is its similarity with the final bound complex state, was based in accordance to the CAPRI (Critical Assessment of PRediction of Interactions) criteria [24], on three quantities: fraction of native contacts (fnat), interface root mean square deviation (IRMSD) and ligand root mean square deviation (LRMSD). These values are used to classify the conformations as incorrect (fnat < 0.1 or (LRMSD > 10Å and IRMSD > 4Å)), acceptable ((fnat ≥ 0.3 and LRMSD > 5Å nd IRMSD > 2Å) or ((fnat ≥ 0.1 and fnat < 0.3) and (LRMSD ≤ 10Åor IRMSD ≤ 4Å))), medium quality ((fnat ≥ 0.5 and LRMSD > 1Å and IRMSD > 1Å) or ((fnat ≥ 0.3 and fnat < 0.5) and (LRMSD ≤ 5Å or IRMSD ≤ 2Å))) or high quality (fnat ≥ 0.5 and (LRMSD ≤ 1Å or IRMSD ≤ 1Å)), relative to the conformation of the native bound complex (i.e. the conformation at the bottom of the binding funnel).


RaTrav: a tool for calculating mean first-passage times on biochemical networks.

Torchala M, Chelminiak P, Kurzynski M, Bates PA - BMC Syst Biol (2013)

The true positive binding funnel. The true positive binding funnel for the vitamin D-binding protein/actin complex (1KXP [22]) generated with the SwarmDock Server [20] by docking unbound receptor/ligand pair included in Benchmark 4.0 [25]. The protein-protein conformational states are numbered from ID 0 to 31. The letter indicates the quality of solution in accordance to CAPRI [24] classification as: M – medium quality (blue), A – acceptable quality (green), I – incorrect solution (red). Favourable paths are marked in red. Figure created with Gephi [26] based on Example1/1KXP.gml file (Additional file 1). See text for more details.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4222613&req=5

Figure 2: The true positive binding funnel. The true positive binding funnel for the vitamin D-binding protein/actin complex (1KXP [22]) generated with the SwarmDock Server [20] by docking unbound receptor/ligand pair included in Benchmark 4.0 [25]. The protein-protein conformational states are numbered from ID 0 to 31. The letter indicates the quality of solution in accordance to CAPRI [24] classification as: M – medium quality (blue), A – acceptable quality (green), I – incorrect solution (red). Favourable paths are marked in red. Figure created with Gephi [26] based on Example1/1KXP.gml file (Additional file 1). See text for more details.
Mentions: The initial network of 32 conformational states, generated by the SwarmDock server, is depicted in Figure 2. The assigned quality of each state, that is its similarity with the final bound complex state, was based in accordance to the CAPRI (Critical Assessment of PRediction of Interactions) criteria [24], on three quantities: fraction of native contacts (fnat), interface root mean square deviation (IRMSD) and ligand root mean square deviation (LRMSD). These values are used to classify the conformations as incorrect (fnat < 0.1 or (LRMSD > 10Å and IRMSD > 4Å)), acceptable ((fnat ≥ 0.3 and LRMSD > 5Å nd IRMSD > 2Å) or ((fnat ≥ 0.1 and fnat < 0.3) and (LRMSD ≤ 10Åor IRMSD ≤ 4Å))), medium quality ((fnat ≥ 0.5 and LRMSD > 1Å and IRMSD > 1Å) or ((fnat ≥ 0.3 and fnat < 0.5) and (LRMSD ≤ 5Å or IRMSD ≤ 2Å))) or high quality (fnat ≥ 0.5 and (LRMSD ≤ 1Å or IRMSD ≤ 1Å)), relative to the conformation of the native bound complex (i.e. the conformation at the bottom of the binding funnel).

Bottom Line: The RaTrav tool can then be applied in order to compute desired MFPTs.For the provided examples, we were able to find the favourable binding path within a protein-protein docking funnel and to calculate the degree of coupling for two chemical reactions catalysed simultaneously by the same protein enzyme.However, the list of possible applications is much wider.

View Article: PubMed Central - HTML - PubMed

Affiliation: Biomolecular Modelling Laboratory, Cancer Research UK London Research Institute, 44 Lincoln's Inn Fields, London WC2A 3LY, UK. Paul.Bates@cancer.org.uk.

ABSTRACT

Background: The concept of mean first-passage times (MFPTs) occupies an important place in the theory of stochastic processes, with the methods of their calculation being equally important in theoretical physics, chemistry and biology. We present here a software tool designed to support computational biology studies where Markovian dynamics takes place and MFPTs between initial and single or multiple final states in network-like systems are used. Two methods are made available for which their efficiency is strongly dependent on the topology of the defined network: the combinatorial Hill technique and the Monte Carlo simulation method.

Results: After a brief introduction to RaTrav, we highlight the utility of MFPT calculations by providing two examples (accompanied by Additional file 1) where they are deemed to be of importance: analysis of a protein-protein docking funnel and interpretation of the free energy transduction between two coupled enzymatic reactions controlled by the dynamics of transition between enzyme conformational states.

Conclusions: RaTrav is a versatile and easy to use software tool for calculating MFPTs across biochemical networks. The user simply prepares a text file with the structure of a given network, along with some additional basic parameters such as transition probabilities, waiting probabilities (if any) and local times (weights of edges), which define explicitly the stochastic dynamics on the network. The RaTrav tool can then be applied in order to compute desired MFPTs. For the provided examples, we were able to find the favourable binding path within a protein-protein docking funnel and to calculate the degree of coupling for two chemical reactions catalysed simultaneously by the same protein enzyme. However, the list of possible applications is much wider.

Show MeSH
Related in: MedlinePlus