Disentangling the Contribution of Each Descriptive Characteristic of Every Single Mutation to Its Functional Effects.

Mutational effects predictions continue to improve in accuracy as advanced artificial intelligence (AI) algorithms are trained on exhaustive experimental data. The next natural questions to ask are if it is possible to gain insights into which attribute of the mutation contributes how much to the mutational effects and if one can develop universal rules for mapping the descriptors to mutational effects. In this work, we mainly address the former aspect using a framework of interpretable AI. Relations between the physicochemical descriptors and their contributions to the mutational effects are extracted by analyzing the data on 29,832 variants from eight systematic deep mutational scan studies. An opposite trend in the dependence of fitness and solubility on the distance of the amino acid from the catalytic sites could be extracted and quantified. The dependence of the mutational effect contributions on the position-specific scoring matrix (PSSM) score for the amino acid after mutation or the BLOSUM score of the substitution showed universal trends. Our attempts in the present work to explain the quantitative differences in the dependence on conservation and SASA across proteins were not successful. The work nevertheless brings transparency into the predictions and development of rules, and will hopefully lead to empirically uncovering the universalities among these rules.

Modelling a pandemic with asymptomatic patients, impact of lockdown and herd immunity, with applications to SARS-CoV-2.

The SARS-CoV-2 is a type of coronavirus that has caused the pandemic known as the Coronavirus Disease of 2019, or COVID-19. In traditional epidemiological models such as SEIR (Susceptible, Exposed, Infected, Removed), the exposed group E does not infect the susceptible group S. A distinguishing feature of COVID-19 is that, unlike with previous viral diseases, there is a distinct "asymptomatic" group A, which does not show any symptoms, but can nevertheless infect others, at the same rate as infected symptomatic patients. This situation is captured in a model known as SAIR (Susceptible, Asymptomatic, Infected, Removed), introduced in Robinson and Stillianakis (2013). The dynamical behavior of the SAIR model is quite different from that of the SEIR model. In this paper, we use Lyapunov theory to establish the global asymptotic stabililty of the SAIR model, both without and with vital dynamics. Then we develop compartmental SAIR models to cater to the migration of population across geographic regions, and once again establish global asymptotic stability. Next, we go beyond long-term asymptotic analysis and present methods for estimating the parameters in the SAIR model. We apply these estimation methods to data from several countries including India, and demonstrate that the predicted trajectories of the disease closely match actual data. We show that "herd immunity" (defined as the time when the number of infected persons is maximum) can be achieved when the total of infected, symptomatic and asymptomatic persons is as low as 25% of the population. Previous estimates are typically 50% or higher. We also conclude that "lockdown" as a way of greatly reducing inter-personal contact has been very effective in checking the progress of the disease.

CMOST: an open-source framework for the microsimulation of colorectal cancer screening strategies.

BACKGROUND: Colorectal cancer (CRC) is a leading cause of cancer-related mortality. CRC incidence and mortality can be reduced by several screening strategies, including colonoscopy, but randomized CRC prevention trials face significant obstacles such as the need for large study populations with long follow-up. Therefore, CRC screening strategies will likely be designed and optimized based on computer simulations. Several computational microsimulation tools have been reported for estimating efficiency and cost-effectiveness of CRC prevention. However, none of these tools is publicly available. There is a need for an open source framework to answer practical questions including testing of new screening interventions and adapting findings to local conditions. METHODS: We developed and implemented a new microsimulation model, Colon Modeling Open Source Tool (CMOST), for modeling the natural history of CRC, simulating the effects of CRC screening interventions, and calculating the resulting costs. CMOST facilitates automated parameter calibration against epidemiological adenoma prevalence and CRC incidence data. RESULTS: Predictions of CMOST were highly similar compared to a large endoscopic CRC prevention study as well as predictions of existing microsimulation models. We applied CMOST to calculate the optimal timing of a screening colonoscopy. CRC incidence and mortality are reduced most efficiently by a colonoscopy between the ages of 56 and 59; while discounted life years gained (LYG) is maximal at 49-50 years. With a dwell time of 13 years, the most cost-effective screening is at 59 years, at $17,211 discounted USD per LYG. While cost-efficiency varied according to dwell time it did not influence the optimal time point of screening interventions within the tested range. CONCLUSIONS: Predictions of CMOST are highly similar compared to a randomized CRC prevention trial as well as those of other microsimulation tools. This open source tool will enable health-economics analyses in for various countries, health-care scenarios and CRC prevention strategies. CMOST is freely available under the GNU General Public License at https://gitlab.com/misselwb/CMOST.

Product Release Pathways in Human and Plasmodium falciparum Phosphoribosyltransferase.

Atomistic molecular dynamics (MD) simulations coupled with the metadynamics technique were carried out to delineate the product (PPi.2Mg and IMP) release mechanisms from the active site of both human (Hs) and Plasmodium falciparum (Pf) hypoxanthine-guanine-(xanthine) phosphoribosyltransferase (HG(X)PRT). An early movement of PPi.2Mg from its binding site has been observed. The swinging motion of the Asp side chain (D134/D145) in the binding pocket facilitates the detachment of IMP, which triggers the opening of flexible loop II, the gateway to the bulk solvent. In PfHGXPRT, PPi.2Mg and IMP are seen to be released via the same path in all of the biased MD simulations. In HsHGPRT too, the product molecules follow similar routes from the active site; however, an alternate but minor escape route for PPi.2Mg has been observed in the human enzyme. Tyr 104 and Phe 186 in HsHGPRT and Tyr 116 and Phe 197 in PfHGXPRT are the key residues that mediate the release of IMP, whereas the motion of PPi.2Mg away from the reaction center is guided by the negatively charged Asp and Glu and a few positively charged residues (Lys and Arg) that line the product release channels. Mutations of a few key residues present in loop II of Trypanosoma cruzi (Tc) HGPRT have been shown to reduce the catalytic efficiency of the enzyme. Herein, in silico mutation of corresponding residues in loop II of HsHGPRT and PfHGXPRT resulted in partial opening of the flexible loop (loop II), thus exposing the active site to bulk water, which offers a rationale for the reduced catalytic activity of these two mutant enzymes. Investigations of the product release from these HsHGPRT and PfHGXPRT mutants delineate the role of these important residues in the enzymatic turnover.

Free-energy landscape of protein oligomerization from atomistic simulations.

In the realm of protein-protein interactions, the assembly process of homooligomers plays a fundamental role because the majority of proteins fall into this category. A comprehensive understanding of this multistep process requires the characterization of the driving molecular interactions and the transient intermediate species. The latter are often short-lived and thus remain elusive to most experimental investigations. Molecular simulations provide a unique tool to shed light onto these complex processes complementing experimental data. Here we combine advanced sampling techniques, such as metadynamics and parallel tempering, to characterize the oligomerization landscape of fibritin foldon domain. This system is an evolutionarily optimized trimerization motif that represents an ideal model for experimental and computational mechanistic studies. Our results are fully consistent with previous experimental nuclear magnetic resonance and kinetic data, but they provide a unique insight into fibritin foldon assembly. In particular, our simulations unveil the role of nonspecific interactions and suggest that an interplay between thermodynamic bias toward native structure and residual conformational disorder may provide a kinetic advantage.

On the recombination of hydronium and hydroxide ions in water.

The recombination of hydronium and hydroxide ions following water ionization is one of the most fundamental processes determining the pH of water. The neutralization step once the solvated ions are in close proximity is phenomenologically understood to be fast, but the molecular mechanism has not been directly probed by experiments. We elucidate the mechanism of recombination in liquid water with ab initio molecular dynamics simulations, and it emerges as quite different from the conventional view of the Grotthuss mechanism. The neutralization event involves a collective compression of the water-wire bridging the ions, which occurs in approximately 0.5 ps, triggering a concerted triple jump of the protons. This process leaves the neutralized hydroxide in a hypercoordinated state, with the implications that enhanced collective compressions of several water molecules around similarly hypercoordinated states are likely to serve as nucleation events for the autoionization of liquid water.

Optimal timing of a colonoscopy screening schedule depends on adenoma detection, adenoma risk, adherence to screening and the screening objective: A microsimulation study.

Colonoscopy-based screening provides protection against colorectal cancer (CRC), but the optimal starting age and time intervals of screening colonoscopies are unknown. We aimed to determine an optimal screening schedule for the US population and its dependencies on the objective of screening (life years gained or incidence, mortality, or cost reduction) and the setting in which screening is performed. We used our established open-source microsimulation model CMOST to calculate optimized colonoscopy schedules with one, two, three or four screening colonoscopies between 20 and 90 years of age. A single screening colonoscopy was most effective in reducing life years lost from CRC when performed at 55 years of age. Two, three and four screening colonoscopy schedules saved a maximum number of life years when performed between 49-64 years; 44-69 years; and 40-72 years; respectively. However, for maximum incidence and mortality reduction, screening colonoscopies needed to be scheduled 4-8 years later in life. The optimum was also influenced by adenoma detection efficiency with lower values for these parameters favoring a later starting age of screening. Low adherence to screening consistently favored a later start and an earlier end of screening. In a personalized approach, optimal screening would start earlier for high-risk patients and later for low-risk individuals. In conclusion, our microsimulation-based approach supports colonoscopy screening schedule between 45 and 75 years of age but the precise timing depends on the objective of screening, as well as assumptions regarding individual CRC risk, efficiency of adenoma detection during colonoscopy and adherence to screening.

Using Atomistic Simulations to Explore the Role of Methylation and ATP in Chemotaxis Signal Transduction.

A bacterial chemotaxis mechanism is activated when nutrients bind to surface receptors. The sequence of intra- and interprotein events in this signal cascade from the receptors to the eventual molecular motors has been clearly identified. However, the atomistic details remain elusive, as in general may be expected of intraprotein signal transduction pathways, especially when fibrillar proteins are involved. We performed atomistic calculations of the methyl accepting chemoprotein (MCP)-CheA-CheW multidomain complex from Escherichia coli, simulating the methylated and unmethylated conditions in the chemoreceptors and the ATP-bound and apo conditions of the CheA. Our results indicate that these atomistic simulations, especially with one of the two force fields we tried, capture several relevant features of the downstream effects, such as the methylation favoring an intermediate structure that is more toward a dipped state and increases the chance of ATP hydrolysis. The results thus suggest the sensitivity of the model to reflect the nutrient signal response, a nontrivial validation considering the complexity of the system, encouraging even more detailed studies on the thermodynamic quantification of the effects and the identification of the signaling networks.

Early Bioinformatic Implication of Triacidic Amino Acid Motifs in Autophagy-Dependent Unconventional Secretion of Mammalian Proteins.

Several proteins are secreted outside the cell, and in many cases, they may be identified by a characteristic signal peptide. However, more and more studies point to the evidence for an "unconventional" secretion, where proteins without a hitherto unknown signal are secreted, possibly in conditions of starvation. In this work, we analyse a set of 202 RNA binding mammalian proteins, whose unconventional secretion has recently been established. Analysis of these proteins secreted by LC3 mediation, the largest unconventionally secreted dataset to our knowledge, identifies the role of KKX motif as well as triacidic amino acid motif in unconventional secretion, the latter being an extension of the recent implicated diacidic amino acid motif. Further data analysis evolves a hypothesis on the sequence or structural proximity of the triacidic or KKX motifs to the LC3 interacting region, and a phosphorylatable amino acid such as serine as a statistically significant feature among these unconventionally secreted proteins. This hypothesis, although needs to be validated in experiments that challenge the specific details of each of these aspects, appears to be one of the early steps in defining what may be a plausible signal for unconventional protein secretion.

Insights on the role of (dis)order from protein-protein interaction linear free-energy relationships.

Protein-protein interactions (PPIs) are remarkably diverse and form the basis for various cellular functions. PPIs can be classified as ordered or disordered; the disordered ones do not have a well-defined structure prior to association, which is an exception to the conventional structure-function relationship. The occurrence of disordered proteins in functional roles is not explained by the conventional structure-function paradigm, and at present there is no clear understanding of the differences between the natures of these two PPIs. In this work, we studied the relationship between the kinetics and thermodynamics in PPIs to provide insights into the latter, with possible implications for the former. Analyzing the experimental data for various protein complexes, we found linear free-energy behavior with a striking kinetic difference between these two types of interactions. Binding affinities of (dis)ordered proteins are correlated with their (association) dissociation rates. Our observation, combined with the correspondence between biological activity and affinity, suggests that selection pressure on the dissociation or association kinetics in a functional context necessitates the presence of (dis)order in the structure.

Probing the mechanism of pH-induced large-scale conformational changes in dengue virus envelope protein using atomistic simulations.

One of the key steps in the infection of the cell by dengue virus is a pH-induced conformational change of the viral envelope proteins. These envelope proteins undergo a rearrangement from a dimer to a trimer, with large conformational changes in the monomeric unit. In this article, metadynamics simulations were used to enable us to understand the mechanism of these large-scale changes in the monomer. By using all-atom, explicit solvent simulations of the monomers, the stability of the protein structure is studied under low and high pH conditions. Free energy profiles obtained along appropriate collective coordinates demonstrate that pH affects the domain interface in both the conformations of E monomer, stabilizing one and destabilizing the other. These simulations suggest a mechanism with an intermediate detached state between the two monomeric structures. Using further analysis, we comment on the key residue interactions responsible for the instability and the pH-sensing role of a histidine that could not otherwise be studied experimentally. The insights gained from this study and methodology can be extended for studying similar mechanisms in the E proteins of the other members of class II flavivirus family.

Toward Developing Intuitive Rules for Protein Variant Effect Prediction Using Deep Mutational Scanning Data.

Protein structure and function can be severely altered by even a single amino acid mutation. Predictions of mutational effects using extensive artificial intelligence (AI)-based models, although accurate, remain as enigmatic as the experimental observations in terms of improving intuitions about the contributions of various factors. Inspired by Lipinski's rules for drug-likeness, we devise simple thresholding criteria on five different descriptors such as conservation, which have so far been limited to qualitative interpretations such as high conservation implies high mutational effect. We analyze systematic deep mutational scanning data of all possible single amino acid substitutions on seven proteins (25153 mutations) to first define these thresholds and then to evaluate the scope and limits of the predictions. At this stage, the approach allows us to comment easily and with a low error rate on the subset of mutations classified as neutral or deleterious by all of the descriptors. We hope that complementary to the accurate AI predictions, these thresholding rules or their subsequent modifications will serve the purpose of codifying the knowledge about the effects of mutations.

Minimal and adaptive numerical strategy for critical resource planning in a pandemic.

Current epidemiological models can in principle model the temporal evolution of a pandemic. However, any such model will rely on parameters that are unknown, which in practice are estimated using stochastic and poorly measured quantities. As a result, an early prediction of the long-term evolution of a pandemic will quickly lose relevance, while a late model will be too late to be useful for disaster management. Unless a model is designed to be adaptive, it is bound either to lose relevance over time, or lose trust and thus not have a second chance for retraining. We propose a strategy for estimating the number of infections and the number of deaths, that does away with time-series modeling, and instead makes use of a "phase portrait approach." We demonstrate that, with this approach, there is a universality to the evolution of the disease across countries, that can then be used to make reliable predictions. These same models can also be used to plan the requirements for critical resources during the pandemic. The approach is designed for simplicity of interpretation, and adaptivity over time. Using our model, we predict the number of infections and deaths in Italy and New York State, based on an adaptive algorithm which uses early available data, and show that our predictions closely match the actual outcomes. We also carry out a similar exercise for India, where in addition to projecting the number of infections and deaths, we also project the expected range of critical resource requirements for hospitalizations in a location.

Modelling the COVID-19 Pandemic: Asymptomatic Patients, Lockdown and Herd Immunity.

The SARS-Cov-2 is a type of coronavirus that has caused the COVID-19 pandemic. In traditional epidemiological models such as SEIR (Susceptible, Exposed, Infected, Removed), the exposed group E does not infect the susceptible group S. A distinguishing feature of COVID-19 is that, unlike with previous viruses, there is a distinct "asymptomatic" group A, who do not show any symptoms, but can nevertheless infect others, at the same rate as infected patients. This situation is captured in a model known as SAIR (Susceptible, Asymptomatic, Infected, Removed), introduced in Robinson and Stilianakis (2013). The dynamical behavior of the SAIR model is quite different from that of the SEIR model. In this paper, we use Lyapunov theory to establish the global asymptotic stabiilty of the SAIR model. Next, we present methods for estimating the parameters in the SAIR model. We apply these estimation methods to data from several countries including India, and show that the predicted trajectories of the disease closely match actual data.

Estimating the herd immunity threshold by accounting for the hidden asymptomatics using a COVID-19 specific model.

A quantitative COVID-19 model that incorporates hidden asymptomatic patients is developed, and an analytic solution in parametric form is given. The model incorporates the impact of lock-down and resulting spatial migration of population due to announcement of lock-down. A method is presented for estimating the model parameters from real-world data, and it is shown that the various phases in the observed epidemiological data are captured well. It is shown that increase of infections slows down and herd immunity is achieved when active symptomatic patients are 10-25% of the population for the four countries we studied. Finally, a method for estimating the number of asymptomatic patients, who have been the key hidden link in the spread of the infections, is presented.

Statistical characteristics of amino acid covariance as possible descriptors of viral genomic complexity.

At the sequence level it is hard to describe the complexity of viruses which allows them to challenge host immune system, some for a few weeks and others up to a complete compromise. Paradoxically, viral genomes are both complex and simple. Complex because amino acid mutation rates are very high, and yet viruses remain functional. Simple because they have barely around 10 types of proteins, so viral protein-protein interaction networks are not insightful. In this work we use fine-grained amino acid level information and their evolutionary characteristics obtained from large-scale genomic data to develop a statistical panel, towards the goal of developing quantitative descriptors for the biological complexity of viruses. Networks were constructed from pairwise covariation of amino acids and were statistically analyzed. Three differentiating factors arise: predominantly intra- vs inter-protein covariance relations, the nature of the node degree distribution and network density. Interestingly, the covariance relations were primarily intra-protein in avian influenza and inter-protein in HIV. The degree distributions showed two universality classes: a power-law with exponent -1 in HIV and avian-influenza, random behavior in human flu and dengue. The calculated covariance network density correlates well with the mortality strengths of viruses on the viral-Richter scale. These observations suggest the potential utility of the statistical metrics for describing the covariance patterns in viruses. Our host-virus interaction analysis point to the possibility that host proteins which can interact with multiple viral proteins may be responsible for shaping the inter-protein covariance relations. With the available data, it appears that network density might be a surrogate for the virus Richter scale, however the hypothesis needs a re-examination when large scale complete genome data for more viruses becomes available.

Computational screening of antimicrobial peptides for Acinetobacter baumannii.

Acinetobacter baumannii, has been developing resistance to even the last line of drugs. Antimicrobial peptides (AMPs) to which bacteria do not develop resistance easily may be the last hope. A few independent experimental studies have designed and studied the activity of AMPs on A. baumannii, however the number of such studies are still limited. With the goal of developing a rational approach to the screening of AMPs against A. baumannii, we carefully curated the drug activity data from 75 cationic AMPs, all measured with a similar protocol, and on the same ATCC 19606 strain. A quantitative model developed and validated with a part of the data. While the model may be used for predicting the activity of any designed AMPs, in this work, we perform an in silico screening for the entire database of naturally occurring AMPs, to provide a rational guidance in this urgently needed drug development.

Dielectric dispersion interpretation of single enzyme dynamic disorder, spectral diffusion, and radiative fluorescence lifetime.

A formulation based on measurable dielectric dispersion of enzymes is developed to estimate fluctuations in electrostatic interaction energy on time scales as long as milliseconds to seconds at a local site in enzymes. Several single molecule experimental obsevations occur on this time scale, currently unreachable by real time computational trajectory simulations. We compare the experimental results on the autocorrelation function of the fluctuations of catalysis rate with the calculations using the dielectric dispersion formulation. We also discuss the autocorrelation functions of the fluorescence lifetime and of spectral diffusion. We use a previously derived relation between the observables and the electric field fluctuations and calculate the latter using dielectric dispersion data for the proteins and the Onsager regression hypothesis.

Exploring the context of diacidic motif DE as a signal for unconventional protein secretion in eukaryotic proteins.

Unconventional protein secretion (UPS) is an important phenomenon with fundamental implications to cargo export. How eukaryotic proteins transported by UPS are recognized without a conventional signal peptide has been an open question. It was recently observed that a diacidic amino acid motif (ASP-GLU or DE) is necessary for the secretion of superoxide dismutase 1 (SOD1) from yeast under nutrient starvation. Taking cue from this discovery, we explore the hypothesis of whether the diacidic motif DE, which can occur fairly ubiquitously, along with its context, can be a generic signal for unconventional secretion of proteins. Four different contexts were evaluated: a physical context encompassing the structural order and charge signature in the neighbourhood of DE, two signalling contexts reflecting the presence of either a phosphorylatable amino acid ('X' in XDE, DXE, DEX) or an LC3 interacting region (LIR) which can trigger autophagy and a co-evolutionary constraint relative to other amino acids in the protein interpreted by examining sequences across different species. Among the 100 proteins we curated from different physiological or pathological conditions, we observe a pattern in the unconventional secretion of heat shock proteins in the cancer secretome, where DE in an ordered structural region has higher odds of being a UPS signal.

Accuracy of current all-atom force-fields in modeling protein disordered states.

Molecular Dynamics (MD) plays a fundamental role in characterizing protein disordered states that are emerging as crucial actors in many biological processes. Here we assess the accuracy of three current force-fields in modeling disordered peptides by combining enhanced-sampling MD simulations with NMR data. These force-fields generate significantly different conformational ensembles, and AMBER03w [ Best and Mittal J. Phys. Chem. B 2010 , 114 , 14916 - 14923 ] provides the best agreement with experiments, which is further improved by adding the ILDN corrections [ Lindorff-Larsen et al. Proteins 2010 , 78 , 1950 - 1958 ].