From Proteins to Ligands: Decoding Deep Learning Methods for Binding Affinity Prediction.

Accurate in silico prediction of protein-ligand binding affinity is important in the early stages of drug discovery. Deep learning-based methods exist but have yet to overtake more conventional methods such as giga-docking largely due to their lack of generalizability. To improve generalizability, we need to understand what these models learn from input protein and ligand data. We systematically investigated a sequence-based deep learning framework to assess the impact of protein and ligand encodings on predicting binding affinities for commonly used kinase data sets. The role of proteins is studied using convolutional neural network-based encodings obtained from sequences and graph neural network-based encodings enriched with structural information from contact maps. Ligand-based encodings are generated from graph-neural networks. We test different ligand perturbations by randomizing node and edge properties. For proteins, we make use of 3 different protein contact generation methods (AlphaFold2, Pconsc4, and ESM-1b) and compare these with a random control. Our investigation shows that protein encodings do not substantially impact the binding predictions, with no statistically significant difference in binding affinity for KIBA in the investigated metrics (concordance index, Pearson's R Spearman's Rank, and RMSE). Significant differences are seen for ligand encodings with random ligands and random ligand node properties, suggesting a much bigger reliance on ligand data for the learning tasks. Using different ways to combine protein and ligand encodings did not show a significant change in performance.

Benchmarking Active Learning Protocols for Ligand-Binding Affinity Prediction.

Active learning (AL) has become a powerful tool in computational drug discovery, enabling the identification of top binders from vast molecular libraries. To design a robust AL protocol, it is important to understand the influence of AL parameters, as well as the features of the data sets on the outcomes. We use four affinity data sets for different targets (TYK2, USP7, D2R, Mpro) to systematically evaluate the performance of machine learning models [Gaussian process (GP) model and Chemprop model], sample selection protocols, and the batch size based on metrics describing the overall predictive power of the model (R2, Spearman rank, root-mean-square error) as well as the accurate identification of top 2%/5% binders (Recall, F1 score). Both models have a comparable Recall of top binders on large data sets, but the GP model surpasses the Chemprop model when training data are sparse. A larger initial batch size, especially on diverse data sets, increased the Recall of both models as well as overall correlation metrics. However, for subsequent cycles, smaller batch sizes of 20 or 30 compounds proved to be desirable. Furthermore, adding artificial Gaussian noise to the data up to a certain threshold still allowed the model to identify clusters with top-scoring compounds. However, excessive noise (<1σ) did impact the model's predictive and exploitative capabilities.

Sire: An interoperability engine for prototyping algorithms and exchanging information between molecular simulation programs.

Sire is a Python/C++ library that is used both to prototype new algorithms and as an interoperability engine for exchanging information between molecular simulation programs. It provides a collection of file parsers and information converters that together make it easier to combine and leverage the functionality of many other programs and libraries. This empowers researchers to use sire to write a single script that can, for example, load a molecule from a PDBx/mmCIF file via Gemmi, perform SMARTS searches via RDKit, parameterize molecules using BioSimSpace, run GPU-accelerated molecular dynamics via OpenMM, and then display the resulting dynamics trajectory in a NGLView Jupyter notebook 3D molecular viewer. This functionality is built on by BioSimSpace, which uses sire's molecular information engine to interconvert with programs such as GROMACS, NAMD, Amber, and AmberTools for automated molecular parameterization and the running of molecular dynamics, metadynamics, and alchemical free energy workflows. Sire comes complete with a powerful molecular information search engine, plus trajectory loading and editing, analysis, and energy evaluation engines. This, when combined with an in-built computer algebra system, gives substantial flexibility to researchers to load, search for, edit, and combine molecular information from multiple sources and use that to drive novel algorithms by combining functionality from other programs. Sire is open source (GPL3) and is available via conda and at a free Jupyter notebook server at https://try.openbiosim.org. Sire is supported by the not-for-profit OpenBioSim community interest company.

What geometrically constrained models can tell us about real-world protein contact maps.

The mechanisms by which a protein's 3D structure can be determined based on its amino acid sequence have long been one of the key mysteries of biophysics. Often simplistic models, such as those derived from geometric constraints, capture bulk real-world 3D protein-protein properties well. One approach is using protein contact maps (PCMs) to better understand proteins' properties. In this study, we explore the emergent behaviour of contact maps for different geometrically constrained models and compare them to real-world protein systems. Specifically, we derive an analytical approximation for the distribution of amino acid distances, denoted asP(s), using a mean-field approach based on a geometric constraint model. This approximation is then validated for amino acid distance distributions generated from a 2D and 3D version of the geometrically constrained random interaction model. For real protein data, we show how the analytical approximation can be used to fit amino acid distance distributions of protein chain lengths ofL ≈ 100,L ≈ 200, andL ≈ 300 generated from two different methods of evaluating a PCM, a simple cutoff based method and a shadow map based method. We present evidence that geometric constraints are sufficient to model the amino acid distance distributions of protein chains in bulk and amino acid sequences only play a secondary role, regardless of the definition of the PCM.

SILVR: Guided Diffusion for Molecule Generation.

Computationally generating new synthetically accessible compounds with high affinity and low toxicity is a great challenge in drug design. Machine learning models beyond conventional pharmacophoric methods have shown promise in the generation of novel small-molecule compounds but require significant tuning for a specific protein target. Here, we introduce a method called selective iterative latent variable refinement (SILVR) for conditioning an existing diffusion-based equivariant generative model without retraining. The model allows the generation of new molecules that fit into a binding site of a protein based on fragment hits. We use the SARS-CoV-2 main protease fragments from Diamond XChem that form part of the COVID Moonshot project as a reference dataset for conditioning the molecule generation. The SILVR rate controls the extent of conditioning, and we show that moderate SILVR rates make it possible to generate new molecules of similar shape to the original fragments, meaning that the new molecules fit the binding site without knowledge of the protein. We can also merge up to 3 fragments into a new molecule without affecting the quality of molecules generated by the underlying generative model. Our method is generalizable to any protein target with known fragments and any diffusion-based model for molecule generation.

Clinical definition of acquired resistance to immunotherapy in patients with metastatic non-small-cell lung cancer.

Acquired resistance (AR) to programmed cell death protein 1/programmed death-ligand 1 [PD-(L)1] blockade is frequent in non-small-cell lung cancer (NSCLC), occurring in a majority of initial responders. Patients with AR may have unique properties of persistent antitumor immunity that could be re-harnessed by investigational immunotherapies. The absence of a consistent clinical definition of AR to PD-(L)1 blockade and lack of uniform criteria for ensuing enrollment in clinical trials remains a major barrier to progress; such clinical definitions have advanced biologic and therapeutic discovery. We examine the considerations and potential controversies in developing a patient-level definition of AR in NSCLC treated with PD-(L)1 blockade. Taking into account the specifics of NSCLC biology and corresponding treatment strategies, we propose a practical, clinical definition of AR to PD-(L)1 blockade for use in clinical reports and prospective clinical trials. Patients should meet the following criteria: received treatment that includes PD-(L)1 blockade; experienced objective response on PD-(L)1 blockade (inclusion of a subset of stable disease will require future investigation); have progressive disease occurring within 6 months of last anti-PD-(L)1 antibody treatment or rechallenge with anti-PD-(L)1 antibody in patients not exposed to anti-PD-(L)1 in 6 months.

Implementation of the QUBE Force Field in SOMD for High-Throughput Alchemical Free-Energy Calculations.

The quantum mechanical bespoke (QUBE) force-field approach has been developed to facilitate the automated derivation of potential energy function parameters for modeling protein-ligand binding. To date, the approach has been validated in the context of Monte Carlo simulations of protein-ligand complexes. We describe here the implementation of the QUBE force field in the alchemical free-energy calculation molecular dynamics simulation package SOMD. The implementation is validated by demonstrating the reproducibility of absolute hydration free energies computed with the QUBE force field across the SOMD and GROMACS software packages. We further demonstrate, by way of a case study involving two series of non-nucleoside inhibitors of HIV-1 reverse transcriptase, that the availability of QUBE in a modern simulation package that makes efficient use of graphics processing unit acceleration will facilitate high-throughput alchemical free-energy calculations.

Dynamic Profiling of ß-Coronavirus 3CL Mpro Protease Ligand-Binding Sites.

ß-coronavirus (CoVs) alone has been responsible for three major global outbreaks in the 21st century. The current crisis has led to an urgent requirement to develop therapeutics. Even though a number of vaccines are available, alternative strategies targeting essential viral components are required as a backup against the emergence of lethal viral variants. One such target is the main protease (Mpro) that plays an indispensable role in viral replication. The availability of over 270 Mpro X-ray structures in complex with inhibitors provides unique insights into ligand-protein interactions. Herein, we provide a comprehensive comparison of all nonredundant ligand-binding sites available for SARS-CoV2, SARS-CoV, and MERS-CoV Mpro. Extensive adaptive sampling has been used to investigate structural conservation of ligand-binding sites using Markov state models (MSMs) and compare conformational dynamics employing convolutional variational auto-encoder-based deep learning. Our results indicate that not all ligand-binding sites are dynamically conserved despite high sequence and structural conservation across ß-CoV homologs. This highlights the complexity in targeting all three Mpro enzymes with a single pan inhibitor.

Outcomes with durvalumab by tumour PD-L1 expression in unresectable, stage III non-small-cell lung cancer in the PACIFIC trial.

BACKGROUND: In the PACIFIC trial, durvalumab significantly improved progression-free and overall survival (PFS/OS) versus placebo, with manageable safety, in unresectable, stage III non-small-cell lung cancer (NSCLC) patients without progression after chemoradiotherapy (CRT). We report exploratory analyses of outcomes by tumour cell (TC) programmed death-ligand 1 (PD-L1) expression. PATIENTS AND METHODS: Patients were randomly assigned (2:1) to intravenous durvalumab 10 mg/kg every 2 weeks or placebo ≤12 months, stratified by age, sex, and smoking history, but not PD-L1 status. Where available, pre-CRT samples were tested for PD-L1 expression (immunohistochemistry) and scored at pre-specified (25%) and post hoc (1%) TC cut-offs. Treatment-effect hazard ratios (HRs) were estimated from unstratified Cox proportional hazards models (Kaplan-Meier-estimated medians). RESULTS: In total, 713 patients were randomly assigned, 709 of whom received at least 1 dose of study treatment durvalumab (n = 473) or placebo (n = 236). Some 451 (63%) were PD-L1-assessable: 35%, 65%, 67%, 33%, and 32% had TC ≥25%, <25%, ≥1%, <1%, and 1%-24%, respectively. As of 31 January 2019, median follow-up was 33.3 months. Durvalumab improved PFS versus placebo (primary-analysis data cut-off, 13 February 2017) across all subgroups [HR, 95% confidence interval (CI); medians]: TC ≥25% (0.41, 0.26-0.65; 17.8 versus 3.7 months), <25% (0.59, 0.43-0.82; 16.9 versus 6.9 months), ≥1% (0.46, 0.33-0.64; 17.8 versus 5.6 months), <1% (0.73, 0.48-1.11; 10.7 versus 5.6 months), 1%-24% [0.49, 0.30-0.80; not reached (NR) versus 9.0 months], and unknown (0.59, 0.42-0.83; 14.0 versus 6.4 months). Durvalumab improved OS across most subgroups (31 January 2019 data cut-off; HR, 95% CI; medians): TC ≥ 25% (0.50, 0.30-0.83; NR versus 21.1 months), <25% (0.89, 0.63-1.25; 39.7 versus 37.4 months), ≥1% (0.59, 0.41-0.83; NR versus 29.6 months), 1%-24% (0.67, 0.41-1.10; 43.3 versus 30.5 months), and unknown (0.60, 0.43-0.84; 44.2 versus 23.5 months), but not <1% (1.14, 0.71-1.84; 33.1 versus 45.6 months). Safety was similar across subgroups. CONCLUSIONS: PFS benefit with durvalumab was observed across all subgroups, and OS benefit across all but TC <1%, for which limitations and wide HR CI preclude robust conclusions.

Hybrid Alchemical Free Energy/Machine-Learning Methodology for the Computation of Hydration Free Energies.

A methodology that combines alchemical free energy calculations (FEP) with machine learning (ML) has been developed to compute accurate absolute hydration free energies. The hybrid FEP/ML methodology was trained on a subset of the FreeSolv database and retrospectively shown to outperform most submissions from the SAMPL4 competition. Compared to pure machine-learning approaches, FEP/ML yields more precise estimates of free energies of hydration and requires a fraction of the training set size to outperform standalone FEP calculations. The ML-derived correction terms are further shown to be transferable to a range of related FEP simulation protocols. The approach may be used to inexpensively improve the accuracy of FEP calculations and to flag molecules which will benefit the most from bespoke force field parametrization efforts.

Assessment of Binding Affinity via Alchemical Free-Energy Calculations.

Free-energy calculations have seen increased usage in structure-based drug design. Despite the rising interest, automation of the complex calculations and subsequent analysis of their results are still hampered by the restricted choice of available tools. In this work, an application for automated setup and processing of free-energy calculations is presented. Several sanity checks for assessing the reliability of the calculations were implemented, constituting a distinct advantage over existing open-source tools. The underlying workflow is built on top of the software Sire, SOMD, BioSimSpace, and OpenMM and uses the AMBER 14SB and GAFF2.1 force fields. It was validated on two datasets originally composed by Schrödinger, consisting of 14 protein structures and 220 ligands. Predicted binding affinities were in good agreement with experimental values. For the larger dataset, the average correlation coefficient Rp was 0.70 ± 0.05 and average Kendall's τ was 0.53 ± 0.05, which are broadly comparable to or better than previously reported results using other methods.

Nivolumab versus docetaxel in previously treated advanced non-small-cell lung cancer (CheckMate 017 and CheckMate 057): 3-year update and outcomes in patients with liver metastases.

Background: Long-term data with immune checkpoint inhibitors in non-small-cell lung cancer (NSCLC) are limited. Two phase III trials demonstrated improved overall survival (OS) and a favorable safety profile with the anti-programmed death-1 antibody nivolumab versus docetaxel in patients with previously treated advanced squamous (CheckMate 017) and nonsquamous (CheckMate 057) NSCLC. We report results from ≥3 years' follow-up, including subgroup analyses of patients with liver metastases, who historically have poorer prognosis among patients with NSCLC. Patients and methods: Patients were randomized 1 : 1 to nivolumab (3 mg/kg every 2 weeks) or docetaxel (75 mg/m2 every 3 weeks) until progression or discontinuation. The primary end point of each study was OS. Patients with baseline liver metastases were pooled across studies by treatment for subgroup analyses. Results: After 40.3 months' minimum follow-up in CheckMate 017 and 057, nivolumab continued to show an OS benefit versus docetaxel: estimated 3-year OS rates were 17% [95% confidence interval (CI), 14% to 21%] versus 8% (95% CI, 6% to 11%) in the pooled population with squamous or nonsquamous NSCLC. Nivolumab was generally well tolerated, with no new safety concerns identified. Of 854 randomized patients across both studies, 193 had baseline liver metastases. Nivolumab resulted in improved OS compared with docetaxel in patients with liver metastases (hazard ratio, 0.68; 95% CI, 0.50-0.91), consistent with findings from the overall pooled study population (hazard ratio, 0.70; 95% CI, 0.61-0.81). Rates of treatment-related hepatic adverse events (primarily grade 1-2 liver enzyme elevations) were slightly higher in nivolumab-treated patients with liver metastases (10%) than in the overall pooled population (6%). Conclusions: After 3 years' minimum follow-up, nivolumab continued to demonstrate an OS benefit versus docetaxel in patients with advanced NSCLC. Similarly, nivolumab demonstrated an OS benefit versus docetaxel in patients with liver metastases, and remained well tolerated. Clinical trial registration: CheckMate 017: NCT01642004; CheckMate 057: NCT01673867.

Impact of domain knowledge on blinded predictions of binding energies by alchemical free energy calculations.

The Drug Design Data Resource (D3R) consortium organises blinded challenges to address the latest advances in computational methods for ligand pose prediction, affinity ranking, and free energy calculations. Within the context of the second D3R Grand Challenge several blinded binding free energies predictions were made for two congeneric series of Farsenoid X Receptor (FXR) inhibitors with a semi-automated alchemical free energy calculation workflow featuring FESetup and SOMD software tools. Reasonable performance was observed in retrospective analyses of literature datasets. Nevertheless, blinded predictions on the full D3R datasets were poor due to difficulties encountered with the ranking of compounds that vary in their net-charge. Performance increased for predictions that were restricted to subsets of compounds carrying the same net-charge. Disclosure of X-ray crystallography derived binding modes maintained or improved the correlation with experiment in a subsequent rounds of predictions. The best performing protocols on D3R set1 and set2 were comparable or superior to predictions made on the basis of analysis of literature structure activity relationships (SAR)s only, and comparable or slightly inferior, to the best submissions from other groups.

Blinded predictions of host-guest standard free energies of binding in the SAMPL5 challenge.

In the context of the SAMPL5 blinded challenge standard free energies of binding were predicted for a dataset of 22 small guest molecules and three different host molecules octa-acids (OAH and OAMe) and a cucurbituril (CBC). Three sets of predictions were submitted, each based on different variations of classical molecular dynamics alchemical free energy calculation protocols based on the double annihilation method. The first model (model A) yields a free energy of binding based on computed free energy changes in solvated and host-guest complex phases; the second (model B) adds long range dispersion corrections to the previous result; the third (model C) uses an additional standard state correction term to account for the use of distance restraints during the molecular dynamics simulations. Model C performs the best in terms of mean unsigned error for all guests (MUE [Formula: see text]-95 % confidence interval) for the whole data set and in particular for the octa-acid systems (MUE [Formula: see text]). The overall correlation with experimental data for all models is encouraging ([Formula: see text]). The correlation between experimental and computational free energy of binding ranks as one of the highest with respect to other entries in the challenge. Nonetheless the large MUE for the best performing model highlights systematic errors, and submissions from other groups fared better with respect to this metric.

Blinded predictions of distribution coefficients in the SAMPL5 challenge.

In the context of the SAMPL5 challenge water-cyclohexane distribution coefficients for 53 drug-like molecules were predicted. Four different models based on molecular dynamics free energy calculations were tested. All models initially assumed only one chemical state present in aqueous or organic phases. Model A is based on results from an alchemical annihilation scheme; model B adds a long range correction for the Lennard Jones potentials to model A; model C adds charging free energy corrections; model D applies the charging correction from model C to ionizable species only. Model A and B perform better in terms of mean-unsigned error ([Formula: see text] D units - 95 % confidence interval) and determination coefficient [Formula: see text], while charging corrections lead to poorer results with model D ([Formula: see text] and [Formula: see text]). Because overall errors were large, a retrospective analysis that allowed co-existence of ionisable and neutral species of a molecule in aqueous phase was investigated. This considerably reduced systematic errors ([Formula: see text] and [Formula: see text]). Overall accurate [Formula: see text] predictions for drug-like molecules that may adopt multiple tautomers and charge states proved difficult, indicating a need for methodological advances to enable satisfactory treatment by explicit-solvent molecular simulations.

Blinded predictions of binding modes and energies of HSP90-α ligands for the 2015 D3R grand challenge.

In the framework of the 2015 D3R inaugural grand challenge, blind binding pose and affinity predictions were performed for a set of 180 ligands of the Heat Shock Protein HSP90-α protein, a relevant cancer target. Spectral clustering was used to rapidly identify alternative binding site conformations in publicly available crystallographic HSP90-α structures. Subsequently, multiple docking and scoring protocols employing the software Autodock Vina and rDock were applied to predict binding modes and rank order ligands. Alchemical free energy calculations were performed with the software FESetup and Sire/OpenMM to predict binding affinities for three congeneric series subsets. Some of the protocols used here were ranked among the top submissions according to most of the evaluation metrics. Docking performance was excellent, but the scoring results were disappointing. A critical assessment of the results is reported, as well as suggestions for future similar competitions.

Statistically optimal analysis of state-discretized trajectory data from multiple thermodynamic states.

We propose a discrete transition-based reweighting analysis method (dTRAM) for analyzing configuration-space-discretized simulation trajectories produced at different thermodynamic states (temperatures, Hamiltonians, etc.) dTRAM provides maximum-likelihood estimates of stationary quantities (probabilities, free energies, expectation values) at any thermodynamic state. In contrast to the weighted histogram analysis method (WHAM), dTRAM does not require data to be sampled from global equilibrium, and can thus produce superior estimates for enhanced sampling data such as parallel/simulated tempering, replica exchange, umbrella sampling, or metadynamics. In addition, dTRAM provides optimal estimates of Markov state models (MSMs) from the discretized state-space trajectories at all thermodynamic states. Under suitable conditions, these MSMs can be used to calculate kinetic quantities (e.g., rates, timescales). In the limit of a single thermodynamic state, dTRAM estimates a maximum likelihood reversible MSM, while in the limit of uncorrelated sampling data, dTRAM is identical to WHAM. dTRAM is thus a generalization to both estimators.

Markov State Models: To Optimize or Not to Optimize.

Markov state models (MSM) are a popular statistical method for analyzing the conformational dynamics of proteins including protein folding. With all statistical and machine learning (ML) models, choices must be made about the modeling pipeline that cannot be directly learned from the data. These choices, or hyperparameters, are often evaluated by expert judgment or, in the case of MSMs, by maximizing variational scores such as the VAMP-2 score. Modern ML and statistical pipelines often use automatic hyperparameter selection techniques ranging from the simple, choosing the best score from a random selection of hyperparameters, to the complex, optimization via, e.g., Bayesian optimization. In this work, we ask whether it is possible to automatically select MSM models this way by estimating and analyzing over 16,000,000 observations from over 280,000 estimated MSMs. We find that differences in hyperparameters can change the physical interpretation of the optimization objective, making automatic selection difficult. In addition, we find that enforcing conditions of equilibrium in the VAMP scores can result in inconsistent model selection. However, other parameters that specify the VAMP-2 score (lag time and number of relaxation processes scored) have only a negligible influence on model selection. We suggest that model observables and variational scores should be only a guide to model selection and that a full investigation of the MSM properties should be undertaken when selecting hyperparameters.

Phase I study of peposertib and avelumab with or without palliative radiotherapy in patients with advanced solid tumors.

INTRODUCTION: We report results from a phase I, three-part, dose-escalation study of peposertib, a DNA-dependent protein kinase inhibitor, in combination with avelumab, an immune checkpoint inhibitor, with or without radiotherapy in patients with advanced solid tumors. MATERIALS AND METHODS: Peposertib 100-400 mg twice daily (b.i.d.) or 100-250 mg once daily (q.d.) was administered in combination with avelumab 800 mg every 2 weeks in Part A or avelumab plus radiotherapy (3 Gy/fraction × 10 days) in Part B. Part FE assessed the effect of food on the pharmacokinetics of peposertib plus avelumab. The primary endpoint in Parts A and B was dose-limiting toxicity (DLT). Secondary endpoints were safety, best overall response per RECIST version 1.1, and pharmacokinetics. The recommended phase II dose (RP2D) and maximum tolerated dose (MTD) were determined in Parts A and B. RESULTS: In Part A, peposertib doses administered were 100 mg (n = 4), 200 mg (n = 11), 250 mg (n = 4), 300 mg (n = 6), and 400 mg (n = 4) b.i.d. Of DLT-evaluable patients, one each had DLT at the 250-mg and 300-mg dose levels and three had DLT at the 400-mg b.i.d. dose level. In Part B, peposertib doses administered were 100 mg (n = 3), 150 mg (n = 3), 200 mg (n = 4), and 250 mg (n = 9) q.d.; no DLT was reported in evaluable patients. Peposertib 200 mg b.i.d. plus avelumab and peposertib 250 mg q.d. plus avelumab and radiotherapy were declared as the RP2D/MTD. No objective responses were observed in Part A or B; one patient had a partial response in Part FE. Peposertib exposure was generally dose proportional. CONCLUSIONS: Peposertib doses up to 200 mg b.i.d. in combination with avelumab and up to 250 mg q.d. in combination with avelumab and radiotherapy were tolerable in patients with advanced solid tumors; however, antitumor activity was limited. GOV IDENTIFIER: NCT03724890.

mRNA interactions with disordered regions control protein activity.

The cytoplasm is compartmentalized into different translation environments. mRNAs use their 3'UTRs to localize to distinct cytoplasmic compartments, including TIS granules (TGs). Many transcription factors, including MYC, are translated in TGs. It was shown that translation of proteins in TGs enables the formation of protein complexes that cannot be established when these proteins are translated in the cytosol, but the mechanism is poorly understood. Here we show that MYC protein complexes that involve binding to the intrinsically disordered region (IDR) of MYC are only formed when MYC is translated in TGs. TG-dependent protein complexes require TG-enriched mRNAs for assembly. These mRNAs bind to a new and widespread RNA-binding domain in neutral or negatively charged IDRs in several transcription factors, including MYC. RNA-IDR interaction changes the conformational ensemble of the IDR, enabling the formation of MYC protein complexes that act in the nucleus and control functions that cannot be accomplished by cytosolically-translated MYC. We propose that certain mRNAs have IDR chaperone activity as they control IDR conformations. In addition to post-translational modifications, we found a novel mode of protein activity regulation. Since RNA-IDR interactions are prevalent, we suggest that mRNA-dependent control of protein functional states is widespread.