China
Africa
Explore chapters and articles related to this topic
Combined models of artificial immune systems
Published in Waldemar Wójcik, Andrzej Smolarz, Information Technology in Medical Diagnostics, 2017
V.I. Lytvynenko, W. Wójcik, A. Smolarz, B. Suleimenov, M. Junisbekov
Let us show the generalised stepwise description of the algorithm. Initialisation. Creation (usually by random generation) of the initial population of antibodies AB.Determination of affinity. For every antibody ABj, ABj∊ AB determine its affinity relative to every antigen Agi, Agi∊ AG. Write the result into the matrix of affinities D : D = [|AG| × mAb], and dij = f(Abj, Agi), dij∊D.Clonal selection and propagation. Select from population n of each the best antibodies for every row of the matrix D and place them into separate population of clones ABC,|ABC| = n ·|AG|. It is necessary to generate clones of elements of the population ABC proportionally to their affinity, i.e. the greater it is, the greater number of clones is generated and vice versa.Affinity maturation. Subject to mutation all the clones of a population ABC with the probability inversely proportional to affinities, i.e. the probability of mutation is the greater, the lower is its affinity. Determine new affinity of every antibody ABj, ABj, ∈ ABC similar to item 2 and obtain the matrix of affinities Dc. Select n antibodies from the population ABC, for which the corresponding vector-column of the DC matrix gives the best generalised result of affinity, and transfer them into the population of memory cells MR.Metadynamics. Substitute the worst d antibodies of the population AB by new random individuals.Substitute n antibodies of the population AB by memory cells from MR and pass to item 2 until the stoppage criterion is reached.
Deciphering deamidation and isomerization in therapeutic proteins: Effect of neighboring residue
Published in mAbs, 2022
Flaviyan Jerome Irudayanathan, Jonathan Zarzar, Jasper Lin, Saeed Izadi
To identify the relevant conformations that enable a nucleophilic attack distance, we sampled the free energy space of ψ & χ1 angles in Ace-GGNAG-Nme pentapeptide using Metadynamics simulations. By applying an additive historical bias, Metadynamics pushes the system to explore conformations that are kinetically limited in an equilibrium simulation50. There are six low energy states in the ψ & χ1 space (Figure 6b), of which two regions are coincident with the near attack conformation distance (Figure 6c). With this information, we looked at the correlation between the near attack side-chain conformation and the experimental rates reported in the Adimab dataset. As with the backbone conformation, we looked at the FES from the equilibrium MD simulation. The side chains were considered to be in a near-attack conformation if the minimum energy values in the regions of FES where −100 < χ1< −20 and −60 < ψ < 60 (Figure 2b, region 2), or 20 < χ1 < 100 and ψ < −60 or ψ > 60 (Figure 2b, region 1), were <1 kcal/mol.
CDR-H3 loop ensemble in solution – conformational selection upon antibody binding
Published in mAbs, 2019
Monica L. Fernández-Quintero, Johannes Kraml, Guy Georges, Klaus R. Liedl
To enhance the sampling of the conformational space, well-tempered metadyamics59-61 simulations were performed in GROMACS62,63 with the PLUMED 2 implementation.64 As collective variables, we used a linear combination of sine and cosine of the ψ torsion angles of CDR-H3 and CDR-L3 loop calculated with functions MATHEVAL and COMBINE implemented in PLUMED 2.64 As discussed previously, the ψ torsion angle captures conformational transitions comprehensively.39,40 The decision to include the CDR-L3 loop ψ torsion angles is based on the structural correlation of the CDR-L3 and CDR-H3 loop and the observed improved sampling efficiency.65 The simulations were performed at 300 K in an NpT ensemble. The height of the Gaussian was determined according to minimal distortion of the antibody systems, resulting in a Gaussian height of 10.0 kcal/mol. Gaussian deposition occurred every 1000 steps and a biasfactor of 10 was used. 1 µs metadynamics simulations were performed for each antibody structure. The resulting trajectories were clustered in cpptraj52,66 by using the average linkage hierarchical clustering algorithm with a distance cutoff criterion of 1.5 Å, resulting in a large number of clusters. As the Ferrochelatase antibody was analyzed previously in a different context with a distance cutoff criterion of 1.0 Å, these data were reused.44 The cluster representatives for the systems were equilibrated and simulated for 100 ns using the AMBER1651 simulation package.
Binding affinity in drug design: experimental and computational techniques
Published in Expert Opinion on Drug Discovery, 2019
Visvaldas Kairys, Lina Baranauskiene, Migle Kazlauskiene, Daumantas Matulis, Egidijus Kazlauskas
Among the relatively new enhanced sampling methods metadynamics and its many variants should be certainly mentioned [107,108]. In metadynamics, every n-th MD step a small Gaussian type ‘hill’ (bias potential) is deposited along the chosen internal coordinate. Since the simulation tends to move to low energy regions, the bias potential gradually fills deep energy wells first. At the end of simulation, the deposited bias potential is explored, and the energy vs. coordinate profile is built. Metadynamics based approaches can be used not only to explore ligand binding to the protein but also to build more complex free energy landscapes along various chosen internal coordinates [109,110]. The binding affinity prediction accuracy using metadynamics can reach less than 1 kcal/mol RMSE [111].