JRJoão Marcelo Lamim Ribeiro

João Marcelo Lamim Ribeiro

*This is an algorithmically generated author record based on publicly available information.

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


João Marcelo Lamim Ribeiro has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Top Cited Papers
    Open Access

    Past–future information bottleneck for sampling molecular reaction coordinate simultaneously with thermodynamics and kinetics

    by Yihang Wang, João Marcelo Lamim Ribeiro, Pratyush Tiwary

    The ability to rapidly learn from high-dimensional data to make reliable bets about the future is crucial in many contexts. This could be a fly avoiding predators, or the retina processing gigabytes of data to guide human actions. In this work we draw parallels between these and the efficient sampling of biomolecules with hundreds of thousands of atoms. For this we use the Predictive Information Bottleneck framework used for the first two problems, and re-formulate it for the sampling of biomolecules, especially when plagued with rare events. Our method uses a deep neural network to learn the minimally complex yet most predictive aspects of a given biomolecular trajectory. This information is used to perform iteratively biased simulations that enhance the sampling and directly obtain associated thermodynamic and kinetic information. We demonstrate the method on two test-pieces, studying processes slower than milliseconds, calculating free energies, kinetics and critical mutations.


  • Kinetics of Ligand–Protein Dissociation from All-Atom Simulations: Are We There Yet?

    by João Marcelo Lamim Ribeiro, Sun-Ting Tsai, Debabrata Pramanik, Yihang Wang, Pratyush Tiwary

    Large parallel gains in the development of both computational resources as well as sampling methods have now made it possible to simulate dissociation events in ligand-protein complexes with all--atom resolution. Such encouraging progress, together with the inherent spatiotemporal resolution associated with molecular simulations, has left their use for investigating dissociation processes brimming with potential, both in rational drug design, where it can be an invaluable tool for determining the mechanistic driving forces behind dissociation rate constants, as well as in force-field development, where it can provide a catalog of transient molecular structures on which to refine force-fields. Although much progress has been made in making force-fields more accurate, reducing their error for transient structures along a transition path could yet prove to be a critical development helping to make kinetic predictions much more accurate. In what follows we will provide a state-of-the-art compilation of the enhanced sampling methods based on molecular dynamics (MD) simulations used to investigate the kinetics and mechanisms of ligand-protein dissociation processes. Due to the timescales of such processes being slower than what is accessible using straightforward MD simulations, several ingenious schemes are being devised at a rapid rate to overcome this obstacle. Here we provide an up-to-date compendium of such methods and their achievements/shortcomings in extracting mechanistic insight into ligand-protein dissociation. We conclude with a critical and provocative appraisal attempting to answer the title of this review.


  • Toward Achieving Efficient and Accurate Ligand-Protein Unbinding with Deep Learning and Molecular Dynamics through RAVE

    by João Marcelo Lamim Ribeiro, Pratyush Tiwary

    In this work we demonstrate how to leverage our recent iterative deep learning-all atom molecular dynamics (MD) technique "Reweighted autoencoded variational Bayes for enhanced sampling (RAVE)" (Ribeiro, Bravo, Wang, Tiwary, J. Chem. Phys. 149, 072301 (2018)) for investigating ligand-protein unbinding mechanisms and calculating absolute binding free energies, ΔGb, when plagued with difficult to sample rare events. In order to do so, we introduce a simple but powerful extension to RAVE that allows learning a reaction coordinate expressed as a piecewise function that is linear over all intervals. Such an approach allows us to retain the physical interpretation of a RAVE-derived reaction coordinate while making the method more applicable to a wider range of complex biophysical problems. As we will demonstrate, using as our test-case the slow dissociation of benzene from the L99A variant of lysozyme, the RAVE extension led to observing an unbinding event in 100% of the independent all-atom MD simulations, all within 3-50 nanoseconds for a process that takes on an average close to few hundred milliseconds, which reflects a seven order of magnitude acceleration relative to straightforward MD. Furthermore, we will show that without the use of time-dependent biasing, clear back-and-forth movement between metastable intermediates was achieved during the various simulations, demonstrating the caliber of the RAVE-derived piecewise reaction coordinate and bias potential, which together drive efficient and accurate sampling of the ligand-protein dissociation event. Last, we report the results for ΔGb, which via very short MD simulations, can form a strict lower-bound that is ∽2-3 kcal/mol off from experiments. We believe that RAVE, together with its multi-dimensional extension that we introduce here, will be a useful tool for simulating the slow unbinding process of practical ligand-protein complexes in an automated manner with minimal use of human intuition.


  • Open Access

    Achieving Reversible Ligand-Protein Unbinding with Deep Learning and Molecular Dynamics through RAVE

    by João Marcelo Lamim Ribeiro, Pratyush Tiwary

    In this work we demonstrate how to leverage our recent iterative deep learning–all atom molecular dynamics (MD) technique “Reweighted autoencoded variational Bayes for enhanced sampling (RAVE)” (Ribeiro, Bravo, Wang, Tiwary, J. Chem. Phys. 149, 072301 (2018)) for sampling protein-ligand unbinding mechanisms and calculating absolute binding affinities when plagued with difficult to sample rare events. RAVE iterates between rounds of MD and deep learning, and unlike other enhanced sampling methods, it stands out in simultaneously learning both a low-dimensional physically interpretable reaction coordinate (RC) and associated free energy. Here, we introduce a simple but powerful extension to RAVE which allows learning a position-dependent RC expressed as a superposition of piecewise linear RCs valid in different metastable states. With this approach, we retain the original physical interpretability of a RAVE-derived RC while making it applicable to a wider range of complex systems. We demonstrate how in its multi-dimensional form introduced here, RAVE can efficiently simulate the unbinding of the tightly bound benzene-lysozyme (L99A variant) complex, in all atom-precision and with minimal use of human intuition except for the choice of a larger dictionary of order parameters. These simulations had a 100 % success rate, and took between 3–50 nanoseconds for a process that takes on an average close to few hundred milliseconds, thereby reflecting a seven order of magnitude acceleration relative to straightforward MD. Furthermore, without any time-dependent biasing, the trajectories display clear back–and– forth movement between various metastable intermediates, demonstrating the reliability of the RC and its probability distribution learnt in RAVE. Our binding free energy is in good agreement with other reported simulation results. We thus believe that RAVE, especially in its multi-dimensional variant introduced here, will be a useful tool for simulating the dissociation process of practical biophysical systems with rare events in an automated manner with minimal use of human intuition.

  • Top Cited Papers
    Open Access

    Reweighted autoencoded variational Bayes for enhanced sampling (RAVE)

    by João Marcelo Lamim Ribeiro, Pablo Bravo, Yihang Wang, Pratyush Tiwary

    Here we propose the reweighted autoencoded variational Bayes for enhanced sampling (RAVE) method, a new iterative scheme that uses the deep learning framework of variational autoencoders to enhance sampling in molecular simulations. RAVE involves iterations between molecular simulations and deep learning in order to produce an increasingly accurate probability distribution along a low-dimensional latent space that captures the key features of the molecular simulation trajectory. Using the Kullback-Leibler divergence between this latent space distribution and the distribution of various trial reaction coordinates sampled from the molecular simulation, RAVE determines an optimum, yet nonetheless physically interpretable, reaction coordinate and optimum probability distribution. Both then directly serve as the biasing protocol for a new biased simulation, which is once again fed into the deep learning module with appropriate weights accounting for the bias, the procedure continuing until estimates of desirable thermodynamic observables are converged. Unlike recent methods using deep learning for enhanced sampling purposes, RAVE stands out in that (a) it naturally produces a physically interpretable reaction coordinate, (b) is independent of existing enhanced sampling protocols to enhance the fluctuations along the latent space identified via deep learning, and (c) it provides the ability to easily filter out spurious solutions learned by the deep learning procedure. The usefulness and reliability of RAVE is demonstrated by applying it to model potentials of increasing complexity, including computation of the binding free energy profile for a hydrophobic ligand–substrate system in explicit water with dissociation time of more than 3 min, in computer time at least twenty times less than that needed for umbrella sampling or metadynamics.


  • Reaction mechanism and product branching ratios of the CH + C 3 H 4 reactions: a theoretical study

    by Joao Marcelo Ribeiro, Alexander M. Mebel

    Two ground-state CH radical reactions with the C3H4 isomers allene and methylacetylene occurring along the C4H5 potential energy surface (PES) were studied to probe the reaction mechanisms and final product distributions. The calculations were performed using a CCSD(T)-F12//B2PLYPD3 PES in combination with the 1-D chemical master equation. The reaction between the CH radical and allene was found to lead to exclusive “funneling” of the energized C4H5 intermediates into linear C4H5 configurations before reaching the exit channels, regardless of the specific nature of the initial bimolecular reactive encounter. In the case of the CH radical reaction with methylacetylene, energized C4H5 three-membered ring structures underwent H loss in significant amounts resulting in the production of a cyclic C4H4 methylenecyclopropene product, in accordance with experiments. The theoretical product distribution at room temperature for methylacetylene + CH was ∼35% methylenecyclopropene, ∼36% vinylacetylene, and ∼28% 1,2,3-butatriene, which is in agreement with the available experimental data. The distribution for allene + CH was ∼93% vinylacetylene, ∼4% 1,2,3-butatriene and ∼3% acetylene + vinyl, which overestimates the experimental yield of vinylacetylene and underestimates that of 1,2,3-butatriene by ∼10%. The possible reasons for this slight quantitative deviation of the theoretical results obtained within statistical treatment from the experiment are discussed.


  • Combined Experimental and Computational Study on the Unimolecular Decomposition of JP-8 Jet Fuel Surrogates. I. n-Decane (n-C10H22)

    by Long Zhao, Tao Yang, Ralf I. Kaiser, Tyler P. Troy, Musahid Ahmed, Daniel Belisario-Lara, Joao Marcelo Ribeiro, Alexander M. Mebel

    Exploiting a high temperature chemical reactor, we explored the pyrolysis of helium-seeded n-decane as a surrogate of the n-alkane fraction of Jet Propellant-8 (JP-8) over a temperature range of 1100–1600 K at a pressure of 600 Torr. The nascent products were identified in situ in a supersonic molecular beam via single photon vacuum ultraviolet (VUV) photoionization coupled with a mass spectroscopic analysis of the ions in a reflectron time-of-flight mass spectrometer (ReTOF). Our studies probe, for the first time, the initial reaction products formed in the decomposition of n-decane—including radicals and thermally labile closed-shell species effectively excluding mass growth processes. The present study identified 18 products: molecular hydrogen (H2), C2 to C7 1-alkenes [ethylene (C2H4) to 1-heptene (C7H14)], C1–C3 radicals [methyl (CH3), vinyl (C2H3), ethyl (C2H5), propargyl (C3H3), allyl (C3H5)], small C1–C3 hydrocarbons [methane (CH4), acetylene (C2H2), allene (C3H4), methylacetylene (C3H4)], along with higher-order reaction products [1,3-butadiene (C4H6), 2-butene (C4H8)]. On the basis of electronic structure calculations, n-decane decomposes initially by C–C bond cleavage (excluding the terminal C–C bonds) producing a mixture of alkyl radicals from ethyl to octyl. These alkyl radicals are unstable under the experimental conditions and rapidly dissociate by C–C bond β-scission to split ethylene (C2H4) plus a 1-alkyl radical with the number of carbon atoms reduced by two and 1,4-, 1,5-, 1,6-, or 1,7-H shifts followed by C–C β-scission producing alkenes from propene to 1-octene in combination with smaller 1-alkyl radicals. The higher alkenes become increasingly unstable with rising temperature. When the C–C β-scission continues all the way to the propyl radical (C3H7), it dissociates producing methyl (CH3) plus ethylene (C2H4). Also, at higher temperatures, hydrogen atoms can abstract hydrogen from C10H22 to yield n-decyl radicals, while methyl (CH3) can also abstract hydrogen or recombine with hydrogen to form methane. These n-decyl radicals can decompose via C–C-bond β-scission to C3 to C9 alkenes.


  • Combined Experimental and Computational Study on the Unimolecular Decomposition of JP-8 Jet Fuel Surrogates. II: n-Dodecane (n-C12H26)

    by Long Zhao, Tao Yang, Ralf I. Kaiser, Tyler P. Troy, Musahid Ahmed, Joao Marcelo Ribeiro, Daniel Belisario-Lara, Alexander M. Mebel

    We investigated temperature-dependent products in the pyrolysis of helium-seeded n-dodecane, which represents a surrogate of the n-alkane fraction of Jet Propellant-8 (JP-8) aviation fuel. The experiments were performed in a high temperature chemical reactor over a temperature range of 1200 K to 1600 K at a pressure of 600 Torr, with in situ identification of the nascent products in a supersonic molecular beam using single photon vacuum ultraviolet (VUV) photoionization coupled with the analysis of the ions in a reflectron time-of-flight mass spectrometer (ReTOF). For the first time, the initial decomposition products of n-dodecane—including radicals and thermally labile closed-shell species—were probed in experiments, which effectively exclude mass growth processes. A total of 15 different products were identified, such as molecular hydrogen (H2), C2 to C7 1-alkenes [ethylene (C2H4) to 1-heptene (C7H14)], C1–C3 radicals [methyl (CH3), ethyl (C2H5), allyl (C3H5)], small C1–C3 hydrocarbons [acetylene (C2H2), allene (C3H4), methylacetylene (C3H4)], as well as the reaction products [1,3-butadiene (C4H6), 2-butene (C4H8)] attributed to higher-order processes. Electronic structure calculations carried out at the G3(CCSD,MP2)//B3LYP/6-311G(d,p) level of theory combined with RRKM/master equation of rate constants for relevant reaction steps showed that n-dodecane decomposes initially by a nonterminal C–C bond cleavage and producing a mixture of alkyl radicals from ethyl to decyl with approximately equal branching ratios. The alkyl radicals appear to be unstable under the experimental conditions and to rapidly dissociate either directly by C–C bond β-scission to produce ethylene (C2H4) plus a smaller 1-alkyl radical with the number of carbon atoms diminished by two or via 1,5-, 1,6-, or 1,7- 1,4-, 1,9-, or 1,8-H shifts followed by C–C β-scission producing alkenes from propene to 1-nonene together with smaller 1-alkyl radicals. The stability and hence the branching ratios of higher alkenes decrease as temperature increases. The C–C β-scission continues all the way to the propyl radical (C3H7), which dissociates to methyl (CH3) plus ethylene (C2H4). In addition, at higher temperatures, another mechanism can contribute, in which hydrogen atoms abstract hydrogen from C12H26 producing various n-dodecyl radicals and these radicals then decompose by C–C bond β-scission to C3 to C11 alkenes.


  • Reaction Mechanism and Product Branching Ratios of the CH + C3H6 Reaction: A Theoretical Study

    by Joao Marcelo Ribeiro, Alexander Moiseevich Mebel

    The mechanism of CH(X2Π) reaction with propene has been studied with ab initio CCSD(T)-F12/CBS//B3LYP/6-311G(d,p) calculations of the C4H7 potential energy surface and RRKM/Master Equation calculations of unimolecular rate constants for the various isomerization and dissociation steps available to the C4H7 radicals. Product branching ratios were calculated and were found to strongly depend on the initial chemically activated C4H7 complex formed in a barrierless entrance channel. If the reaction is initiated via either CH addition to the double bond in propene or CH insertions into the terminal sp2 C-H or single C-C bonds, then 1,3-butadiene + H are predicted to be the dominant products, ethene + C2H3 radical are minor but non-negligible products, and a small amount of 1,2-butadiene + H is also produced. The reaction then proceeds through a key CH3CHCH•CH2 intermediate, which loses an H atom to form either 1,3- or 1,2-butadiene or isomerizes to •CH2CH2CHCH2 and then dissociates to ethene + C2H3 radical. If CH inserts into a C-H bond in the CH3 group the •CH2CH2CHCH2 complex is formed directly and then the major reaction products are predicted to be ethene + C2H3 radical and 1,3-butadiene + H. Finally, if CH inserts into the middle sp2 C-H bond, a branched CH3C(•CH2)CH2 complex is produced, which predominantly decomposes to allene + CH3 radical. A comparison of the calculated reaction mechanism with available experimental data indicates that the CH addition entrance channel is favorable, in which case the computed branching ratios are in agreement with the experimental result of Loison and Bergeat (Ref. 13) who measured the H elimination branching ratio of 78+10%. However, the computed branching ratios quantitatively disagree with the experimental data by Trevitt et al. (Ref. 14), who observed a nearly 100% yield of the C4H6 + H products and also larger yields of 1,2-butadiene and 1-butyne than the calculations predict. The deviation of the theoretical results from experiment can be rationalized in terms of dynamical factors, which should favor direct dissociation of the CH3CHCH•CH2 precursor by H loss, especially to 1,2-butadiene, over its isomerization to •CH2CH2CHCH2 followed by the production of ethene + C2H3 radical, while 1-butyne might be formed through secondary H assisted isomerization of 1,2-butadiene. Overall, the calculations corroborate that the CH + C3H6 reaction could be a major source of 1,3-butadiene at low temperature and low pressure conditions in the interstellar medium and planetary atmospheres.