The volume of data available in biology has increased tremendously (Marx, 2013; Stephens et al., 2015), through the emergence of high-throughput experimental technologies, often referred to as -omics, and the development of efficient computational techniques, associated with high-performance computing resources. The Open Access (OA) movement to make research results free and available to anyone (including e.g. the Budapest Open Access Initiative and the Berlin declaration on Open Access to Knowledge) has led to an explosive growth of research data made available by scientists (Wilson et al., 2021). The FAIR (Findable, Accessible, Interoperable and Reusable) principles Wilkinson et al., 2016 have emerged to structure the sharing of these data with the goals of reusing research data and to contribute to the scientific reproducibility. This leads to a world where research data has become widely available and exploitable, and consequently new applications based on artificial intelligence (AI) emerged. One example is AlphaFold (Jumper et al., 2021), which enables the construction of a structural model of any protein from its sequence. However, it is important to be aware that the development of AlphaFold was only possible because of the existence of extremely well annotated and cleaned open databases of protein structures (wwPDB Berman et al., 2003) and sequences (UniProt Consortium, 2022). Similarly, accurate predictions of NMR chemical shifts and chemical-shift-driven structure determination was only made possible via a community-driven collection of NMR data in the Biological Magnetic Resonance Data Bank (Hoch et al., 2023). One can easily imagine novel possibilities of AI and deep learning reusing previous research data in other fields, if that data is curated and made available at a large scale (Fan and Shi, 2022; Mahmud et al., 2021).

Molecular Dynamics (MD) is an example of a well-established research field where simulations give valuable insights into dynamic processes, ranging from biological phenomena to material science (Perilla et al., 2015; Hollingsworth and Dror, 2018; Yoo et al., 2020; Alessandri et al., 2021; Krishna et al., 2021). By unraveling motions at details and timescales invisible to the eye, this well-established technique complements numerous experimental approaches (Bottaro and Lindorff-Larsen, 2018; Marklund and Benesch, 2019; Fawzi et al., 2021). Nowadays, large amounts of MD data could be generated when modelling large molecular systems (Gupta et al., 2022) or when applying biased sampling methods (Hénin et al., 2022). Most of these simulations are performed to decipher specific molecular phenomena, but typically they are only used for a single publication. We have to confess that many of us used to believe that it was not worth the storage to collect all simulations (in particular since all might not have the same quality), but in hindsight this was wrong. Storage is exceptionally cheap compared to the resources used to generate simulations data, and they represent a potential goldmine of information for researchers wanting to reanalyze them (Antila et al., 2021), in particular when modern machine-learning methods are typically limited by the amount of training data. In the era of open and data-driven science, it is critical to render the data generated by MD simulations not only technically available but also practically usable by the scientific community. In this endeavor, discussions started a few years ago (Abraham et al., 2019; Abriata et al., 2020; Merz et al., 2020) and the MD data sharing trend has been accelerated with the effort of the MD community to release simulation results related to the COVID-19 pandemic (Amaro and Mulholland, 2020; Mulholland and Amaro, 2020) in a centralized database (https://covid.bioexcel.eu). Specific databases have also been developed to store sets of simulations related to protein structures (MoDEL: Meyer et al., 2010), membrane proteins in general (MemProtMD: Stansfeld et al., 2015; Newport et al., 2019), G-protein-coupled receptors in particular (GPCRmd: Rodríguez-Espigares et al., 2020), or lipids (Lipidbook:Domański et al., 2010, NMRLipids Databank: Kiirikki et al., 2023).

Albeit previous attempts in the past (Tai et al., 2004; Meyer et al., 2010), there is, as of now, no central data repository that could host all kinds of MD simulation files. This is not only due to the huge volume of data and its heterogeneity, but also because interoperability of the many file formats used adds to the complexity. Thus, faced with the deluge of biosimulation data (Hospital et al., 2020), researchers often share their simulation files in multiple generalist data repositories. This makes it difficult to search and find available data on, for example, a specific protein or a given set of parameters. We are qualifying this amount of scattered data as the dark matter of MD, and we believe it is essential to shed light onto this overlooked but high-potential volume of data. When unlocked, publicly available MD files will gain more visibility. This will help people to access and reuse these data more easily and overall, by making MD simulation data more FAIR (Wilkinson et al., 2016), it will also improve the reproducibility of MD simulations (Elofsson et al., 2019; Porubsky et al., 2020; Bonomi et al., 2019).

In this work, we have employed a search strategy to index scattered MD simulation files deposited in generalist data repositories. With a focus on the files generated by the Gromacs MD software, we performed a proof-of-concept large-scale analysis of publicly available MD data. We revealed the high value of these data and highlighted the different categories of the simulated molecules, as well as the biophysical conditions applied to these systems. Based on these results and our annotations, we proposed a search engine prototype to easily explore this dark matter of MD. Finally, building on this experience, we provide simple guidelines for data sharing to gradually improve the FAIRness of MD data.

With the rise of open science, researchers increasingly share their data and deposit them into generalist data repositories, such as Zenodo (https://zenodo.org), Figshare (https://figshare.com), Open Science Framework (OSF, https://osf.io), and Dryad (https://datadryad.org/). In this first attempt to find out how many files related to MD are deposited in data repositories, we focused our exploration on three major data repositories: Figshare (∼3.3 million files, ∼112 TB of data, as of January 2023), OSF (∼2 million files, as of November 2022) [Figures provided by Figshare and OSF user support teams.], and Zenodo (∼9.9 million files, ∼1.3 PB of data, as of December 2022; Panero and Benito, 2022).

One immediate strategy to index MD simulation files available in data repositories is to perform a text-based Google-like search. For that, one queries these repositories with keywords such as ‘molecular dynamics’ or ‘Gromacs’. Unfortunately, we experienced many false positives with this search strategy. This could be explained by the strong discrepancy we observed in the quantity and quality of metadata (title, description) accompanying datasets and queried in text-based search. For instance, a description text could be composed of a couple of words to more than 1200 words. Metadata is provided by the user depositing the data, with no incentive to issue relevant details to support the understanding of the simulation. For the three data repositories studied, no human curation other by that of the providers is performed when submitting data. It is also worth mentioning that title and description are provided as free-text and do not abide to any controlled vocabulary such as a specific MD ontology.

To circumvent this issue, we developed an original and specific search strategy that we called Explore and Expand ($E{x}^2$) (see Figure 1A and Materials and methods section) and that relies on a combination of file types and keywords queries. In the Explore phase, we searched for files based on their file types (for instance: .xtc, .gro, etc) with MD-related keywords (for instance: ‘molecular dynamics’, ‘Gromacs’, ‘Martini’, etc). Each of these hit files belonged to a dataset, which we further screened in the Expand phase. There, we indexed all files found in a dataset identified in the previous Explore phase with, this time, no restriction to the collected file types (see Figure 1A and details on the data scraping procedure in the Materials and methods section).

Figure 1

Download asset Open asset

Globally, we indexed about 250,000 files and 2000 datasets that represented 14 TB of data deposited between August 2012 and March 2023 (see Table 1). One major difficulty were the numerous files stored in zipped archives, about seven times more than files steadily available in datasets (see Table 1). While this choice is very convenient for depositing the files (as one just needs to provide one big zip file to upload to the data repository server), it hinders the analysis of MD files as data repositories only provide a limited preview of the content of the zip archives and completely inhibits, for example, data streaming for remote analysis and visualization. Files within zip files are not indexed and cannot be searched individually. The use of zip archives also hampers the reusability of MD data, since a specific file cannot be downloaded individually. One has to download the entire zip archive (sometimes with a size up to several gigabytes) to extract the one file of interest.

The first dataset we found related to MD data that has been deposited in August 2012 in Figshare and corresponds to the work of Fuller et al., 2012 (see Table 1) but we may consider the start of more substantial deposition of the MD data to be 2016 with more than 20,000 files deposited, mainly in Figshare (see Figure 1B). While the number of files deposited in Zenodo was first relatively limited, the last few years (2020–2022) saw a steep increase, passing from a few thousands files in 2018 to almost 50,000 files in 2022 (see Figure 1B). In 2018, the number of MD files deposited in OSF was similar to those in the two other data repositories, but did not take off as much as the other data repositories. Zenodo seems to be favored by the MD community since 2019, even though Figshare in 2022 also saw a sharp increase in deposited MD files. The preference for Zenodo could also be explained by the fact that it is a publicly funded repository developed under the European OpenAIRE program and operated by CERN (European Organization For Nuclear Research, 2013). Overall, the trend showed a rise of deposited data with a steep increase in 2022 (Figure 1B). We believe that this trend will continue in future years, which will lead to a greater amount of MD data available. It is thus urgent to deploy a strategy to index this vast amount of data, and to allow the MD community to easily explore and reuse such gigantic resource. The following describes what is already feasible in terms of meta analysis, in particular what types of data are deposited in data repositories and the simulation setup parameters used by MD experts that have deposited their data.

With our $E{x}^2$ strategy (see Figure 1A), we assigned the deposited files to the MD packages: AMBER (Salomon‐Ferrer et al., 2013), DESMOND (Bowers et al., 2006), Gromacs (Berendsen et al., 1995; Abraham et al., 2015), and NAMD/CHARMM (Phillips et al., 2020; Brooks et al., 2009), based on their corresponding file types (see Materials and methods section). In the case of NAMD/CHARMM, file extensions were mostly identical, which prevented us from distinguishing the respective files from these two MD programs. With 87,204 files deposited, the Gromacs program was most represented (see Figure 2A), followed by NAMD/CHARMM, AMBER, and DESMOND. This statistic is limited as it does not consider more specific databases related to a particular MD program. For example, the DE Shaw Research website contains a large amount of simulation data related to SARS-CoV-2 that has been generated using the ANTON supercomputer (https://www.deshawresearch.com/downloads/download_trajectory_sarscov2.cgi/) or other extensively simulated systems of interest to the community. However, this in itself might also serve as a good example, since few automated search strategies will be able to find custom stand-alone web servers as valuable repositories. Here, our goal was not to compare the availability of all data related to each MD program but to give a snapshot of the type of data available at a given time (i.e. March 2023) in generalist data repositories. Interestingly, many files (>133,000) were not directly associated to any MD program (see Figure 2A label ‘Unknown’). We categorized these files based on their extensions (see Figure 2B). While 10% of these files were without file extension (Figure 2B, column none), we found numerous files corresponding to structure coordinates such as .pdb (∼12,000) and .xyz (∼6800) files. We also got images (.tiff files) and graphics (.xvg files). Finally, we found many text files such as .txt, .dat, and .out which can potentially hold details about how simulations were performed. Focusing further on files related to the Gromacs program, being currently most represented in the studied data repositories, we demonstrated in the following present possibilities to retrieve numerous information related to deposited MD simulations.

Figure 2

Download asset Open asset

First, we were interested in what file types researchers deposited and thereby find potentially of great value to share. We therefore quantified the types of files generated by Gromacs (Figure 3A). The most represented file type is the.xtc file (28,559 files, representing 8.6 TB). This compressed (binary) file is used to store the trajectory of an MD simulation and is an important source of information to characterize the evolution of the simulated molecular system as a function of time. It is thus logical to mainly find this type of file shared in data repositories, as it is of great value for reusage and new analyses. Nevertheless, it is not directly readable but needs to be read by a third-party program, such as Gromacs itself, a molecular viewer like VMD (Humphrey et al., 1996) or an analysis library such as MDAnalysis (Gowers et al., 2016; Michaud-Agrawal et al., 2011). In addition, this trajectory file can only be of use in combination with a matching coordinates file, in order to correctly access the dynamics information stored in this file. Thus, as it is, this file is not easily mineable to extract useful information, especially if multiple .xtc and coordinate files are available in one dataset. Interestingly, we found 1406 .trr files, which contain trajectory but also additional information such as velocities, energy of the system, etc. While this file is especially useful in terms of reusability, the large size (can go up to several 100 GB) limits its deposition in most data repositories. For instance, a file cannot usually exceed 50 GB in Zenodo, 20 GB in Figshare (for free accounts) and 5 GB in OSF. Altogether, Gromacs trajectory files represented about 30,000 files in the three explored generalist repositories (34% of Gromacs files). This is a large number in comparison to existing trajectories stored in known databases dedicated to MD with 1700 MD trajectories available in MoDEL, 1737 trajectories (as of November 2022) available in GPCRmd, 5971 (as of January 2022) trajectories available in MemProtMD and 726 trajectories (as of March 2023) available in the NMRLipids Databank. Although fewer in count, these numbers correspond to manually or semi-automatically curated trajectories of specific systems, mostly proteins and lipids. Thus, ∼30,000 MD trajectories available in generalist data repositories may represent a wider spectrum of simulated systems but need to be further analyzed and filtered to separate usable data from less interesting trajectories such as minimization or equilibration runs.

Figure 3

Download asset Open asset

Given the large volume of data represented by .xtc files (see above), we could only scratch the surface of the information stored in these trajectory files by analyzing a subset of 779 .xtc files - one per dataset in which this type of file was found. We were able to get the size of the molecular systems and the number of frames available in these files (Figure 3B). The system size was up to more than one million atoms for a simulation of the TonB protein (Virtanen et al., 2020). The cumulative distribution of the number of frames showed that half of the files contain more than 10,000 frames. This conformational sampling can be very useful for other research fields besides the MD community that study, for instance, protein flexibility or protein engineering where diverse backbones can be of value. We found an .xtc file containing more than 5 million frames, where the authors probe the picosecond–nanosecond dynamics of T4 lysozyme and guide the MD simulation with NMR relaxation data (Kümmerer et al., 2021). Extending this analysis to all 28,559 .xtc files detected would be of great interest for a more holistic view, but this would require an initial step of careful checking and cleaning to be sure that these files are analyzable. Of note, as .xtc files also contain time stamps, it would be interesting to study the relationship between the time and the number of frames to get useful information about the sampling. Nevertheless, this analysis would be possible only for unbiased MD simulations. So, we would need to decipher if the .xtc file is coming from biased or unbiased simulations, which may not be trivial.

These results bring a first explanation on why there is not a single special-purpose repository for MD trajectory files. Databases dedicated to molecular structures such as the Protein Databank (Berman et al., 2000; Kinjo et al., 2017; Armstrong et al., 2020), or even the recent PDB-dev (Burley et al., 2017), designed for integrative models, cannot accept such large-size files, even less if complete trajectories without reducing the number of frames would be uploaded. This would also require implementing extra steps of data curation and quality control. In addition, the size of the IT infrastructure and the human skills required for data curation represents a significant cost that could probably not be supported by a single institution.

Subsequently, our interest shifted towards exploring which systems are being investigated by MD researchers who deposit their files. We found 9718 .gro files which are text files that contain the number of particles and the Cartesian coordinates of the system modelled. By parsing the number of particles and the type of residue, we were able to give an overview of all Gromacs systems deposited (Figure 3C, D). In terms of system size, they ranged from very small - starting with two coarse-grain (CG) particles of graphite (Piskorz et al., 2019), followed by coordinates of a water molecule (3 atoms) (Ivanov et al., 2017), CG model of benzene (3 particles) (Dandekar and Mondal, 2020) and atomistic model of ammonia (4 atoms) (Kelly and Smith, 2020) — to go up to atomistic and coarse-grain systems composed of more than 3 million particles (Duncan et al., 2020; Schaefer and Hummer, 2022; Figure 3C). Interestingly, the system sizes in .gro files exceeded those of the analyzed .xtc files (Figure 3B). Even if we cannot exclude that the limited number of .xtc files analyzed (779 .xtc files selected from 28,559 .xtc files indexed) could explain this discrepancy, an alternate hypothesis is that the size of an .xtc file also depends on the number of frames stored. To reduce the size of .xtc files deposited in data repositories, besides removing some frames, researchers might also remove parts of the system, such as water molecules. As a consequence for reusability, this solvent removal could limit the number of suitable datasets available for researchers interested in re-analysing the simulation with respect to, in this case, water diffusion. While the size of systems extracted from .gro files was homogeneously spread, we observed a clear bump around system sizes of circa 8500 atoms/particles. This enrichment of data could be explained by the deposition of ∼340 .gro files related to the simulation of a peptide translocation through a membrane (Figure 3C; Kabelka et al., 2021). Beyond 1 million particles/atoms, the number of systems is, for the moment, very limited.

We then analyzed residues in .gro files and inferred different types of molecular systems (see Figure 3D). Two of the most represented systems contained lipid molecules. This may be related to NMRLipids initiative (http://nmrlipids.blogspot.com). For several years, this consortium has been actively working on lipid modelling with a strong policy of data sharing and has contributed to share numerous datasets of membrane systems. As illustrated in Figure 3C, a variety of membrane systems, especially membrane proteins, were deposited. This highlights the vitality of this research field, and the will of this community to share their data. We also found numerous systems containing solvated proteins. This type of data, combined with .xtc trajectory files (see above), could be invaluable to describe protein dynamics and potentially train new artificial intelligence models to go beyond the current representation of the static protein structure (Lane, 2023). There was also a good proportion of systems containing nucleic acids alone or in interaction with proteins (1237 systems). At this time, we found only few systems containing carbohydrates that also contained proteins and corresponded to one study to model hyaluronan–CD44 interactions (Vuorio et al., 2017). Maybe a reason for this limited number is that systems containing sugars are often modelled using AMBER force field (Salomon‐Ferrer et al., 2013), in combination with GLYCAM (Kirschner et al., 2008). A future study on the ∼10,200 AMBER files deposited could retrieve more data related to carbohydrate containing systems. Given the current developments to model glycans (Fadda, 2022), we expect to see more deposited systems with carbohydrates in the coming years.

Finally, we found 1029 gro files which did not belong to the categories previously described. These files were mostly related to models of small molecules, or molecules used in organic chemistry (Young et al., 2020) and material science (Piskorz et al., 2019; Zheng et al., 2022) (see central panel, Figure 3D). Several datasets contained lists of small molecules used for calculating free energy of binding (Aldeghi et al., 2016), solubility of molecules (Liu et al., 2016), or osmotic coefficient (Zhu, 2019). Then, we identified models of nanoparticles (Kyrychenko et al., 2012; Pohjolainen et al., 2016), polymers (Sarkar et al., 2020; Karunasena et al., 2021; Gertsen et al., 2020), and drug molecules like EPI-7170, which binds disordered regions of proteins (Zhu et al., 2022). Finally, an interesting case from material sciences was the modelling of the PTEG-1 molecule, an addition of polar triethylene glycol (TEG) onto a fulleropyrrolidine molecule (see central panel, Figure 3D). This molecule was synthesized to improve semiconductors (Jahani et al., 2014). We found several models related to this peculiar molecule and its derivatives, both atomistic (Qiu et al., 2017; Sami et al., 2022) and coarse grained (Alessandri et al., 2020). With a good indexing of data and appropriate metadata to identify modelled molecules, a simple search, which was previously to this study missing, could easily retrieve different models of the same molecule to compare them or to run multi-scale dynamics simulations. Beyond .gro files, we would like to analyze the ensemble of the ∼12,000 .pdb extracted in this study (see Figure 2B) to better characterize the types of molecular structures deposited.

Another important category of deposited files are those containing information about the topology of the simulated molecules, including file extensions such as .itp and .top. Further, they are often the results of long parametrization processes (Vanommeslaeghe and MacKerell, 2012; Souza et al., 2021; Wang et al., 2004) and therefore of significant value for reusability. Based on our analysis, we indexed almost 20,000 topology files which could spare countless efforts to the MD community if these files could be easily found, annotated and reused. Interestingly, the number of .itp files was elevated (13,058 files) with a total size of 2 GB, while there were less .top files (7009 files) with a total size of 17 GB. Thus, .itp files seemed to contain much less information than the .top files. Among the remaining file types, .tpr files contain all the information to potentially directly run a simulation. Here, we found 4987 .tpr files, meaning that it could virtually be possible to rerun almost 5000 simulations without the burden of setting up the system to simulate. Finally, the 3730 .log files are also a source of useful information as it is relatively easy to parse this text file to extract details on how MD simulations were run, such as the version of Gromacs, which command line was used to run the simulation, etc.

Our next step was to gain insight into the parameter settings employed by the MD community, which may aid us in identifying preferences in MD setups and potential necessity for further education to avoid suboptimal or outdated configurations. We therefore analyzed 10,055 .mdp files stored in the different data repositories. These text files contain information regarding the input parameters to run the simulations such as the integrator, the number of steps, the different algorithms for barostat and thermostat, etc. (for more details see: https://manual.gromacs.org/documentation/current/user-guide/mdp-options.html).

We determined the expected simulation time corresponding to the product of two parameters found in .mdp files: the number of steps and the time step. Here, we acknowledge that one can set up a very long simulation time and stop the simulation before the end or, on contrary, use a limited time (especially when calculations are performed on HPC resources with wall-time) and then extend the simulation for a longer duration. Using only the .mdp file, we cannot know if the simulation reached its term. To do so, comparison with an .xtc file from the same dataset may help to answer this specific question. However, in this study, we were interested in MD setup practices, in particular what simulation time researchers would set up their system with - likely in the mindset to reach that ending time. We restricted this analysis to the 4623 .mdp files that used the md or sd integrator, and that have a simulation time above 1 ns. We found that the majority of the .mdp files were used for simulations of 50 ns or less (see Figure 4A). Further, 697 .mdp files with simulations times set-up between 50 ns and 1 µs and 585 .mdp files with simulation time above 1 µs were identified. As analyzing .gro files showed a good proportion of coarse-grained models (Figure 3B, C), we discriminated simulations setups for these two types of models using the time step as a simple cutoff. We considered that a time step greater than 10 fs (i.e. dt = 0.01) corresponded to MD setups for coarse grained models (Ingólfsson et al., 2014). Globally, we found that over all simulations, the setups for atomistic simulations were largely dominant. However, for simulations with a simulation time above 1 µs specifically, coarse-grain simulations represented 86% of all.

Figure 4

Download asset Open asset

We then looked into the combinations of thermostat and barostat (see Figure 4B) from 9199 .mdp files. The main thermostat used is by far the V-rescale (Bussi et al., 2007) often associated with the Parrinello-Rahman barostat (Parrinello and Rahman, 1981). This thermostat was also used with the Berendsen barostat (Berendsen et al., 1984). In a few cases, we observed the use of the V-rescale thermostat with the very recently developed C-rescale barostat (Bernetti and Bussi, 2020). A total of 2021 .mdp files presented neither thermostat nor barostat, which means they would not be used in production runs. This could correspond to setups used for energy minimization, or to add ions to the system (with the genion command), or for molecular mechanics with Poisson–Boltzmann and surface area solvation (MM/PBSA) and molecular mechanics with generalised Born and surface area solvation (MM/GBSA) calculations (Genheden and Ryde, 2015).

Finally, we analyzed the range of starting temperatures used to perform simulations (see Figure 4C). We found a clear peak around the temperatures 298 K - 310 K which corresponds to the range between ambient room (298 K - 25 °C) and physiological (310 K - 37 °C) temperatures. Nevertheless, we also observed lower temperatures, which often relate to studies of specific organic systems or simulations of Lennard-Jones models (Jeon et al., 2016). Interestingly, we noticed the appearance of several pikes at 400 K, 600 K, and 800 K, which were not present before the end of the year 2022. These peaks corresponded to the same study related to the stability of hydrated crystals (Dybeck et al., 2023). Overall, this analysis revealed that a wide range of temperatures have been explored, starting mostly from 100 K and going up to 800 K.

To encourage further analysis of the collected files, we shared our data collection with the community in Zenodo (see Data availability statement). The data scrapping procedure and data analysis is available on GitHub with a detailed documentation. To let researchers having a quick glance and explore this data collection, we created a prototype web application called MDverse data explorer available at https://mdverse.streamlit.app/ and illustrated in Figure 5A. With this web application, it is easy to use keywords and filters to access interesting datasets for all MD engines, as well as .gro and .mdp files. Furthermore, when available, a description of the found data is provided and searchable for keywords (Figure 5A, on the left sidebar). The sets of data found can then be exported as a tab-separated values (.tsv) file for further analysis (Figure 5B).

Figure 5

Download asset Open asset

Data files produced from the data collection and processing are shared in Parquet format in Zenodo. They are freely available under the Creative Commons Attribution 4.0 International license (CC-BY). Python scripts to search and index MD files, and to download and parse .mdp and.gro files are open-source (under the AGPL-3.0 license), freely available on GitHub and archived in Software Heritage (Poulain et al., 2023). A detailed documentation is provided along the scripts to easily reproduce the data collection and processing. Jupyter notebooks used to analyze results and create the figures of this paper are open-source (under the BSD 3-Clause license), freely available on GitHub and archived in Software Heritage (Poulain, 2023). The code of the MDverse data explorer web application is open-source (under the BSD 3-Clause license), freely available on GitHub and archived in Software Heritage (Poulain and Oussaren, 2023).

1. 1. Johanna KST
   2. Mathieu C
   3. Pierre P

   (2023) Zenodo

   MDverse datasets.

   https://doi.org/10.5281/zenodo.7856806

1. 1. Abraham MJ
   2. Murtola T
   3. Schulz R
   4. Páll S
   5. Smith JC
   6. Hess B
   7. Lindahl E

   (2015) GROMACS: High performance molecular simulations through multi-level parallelism from laptops to supercomputers

   SoftwareX 1–2:19–25.

   https://doi.org/10.1016/j.softx.2015.06.001

   • Google Scholar
2. 1. Abraham M
   2. Apostolov R
   3. Barnoud J
   4. Bauer P
   5. Blau C
   6. Bonvin AMJJ
   7. Chavent M
   8. Chodera J
   9. Čondić-Jurkić K
   10. Delemotte L
   11. Grubmüller H
   12. Howard RJ
   13. Jordan EJ
   14. Lindahl E
   15. Ollila OHS
   16. Selent J
   17. Smith DGA
   18. Stansfeld PJ
   19. Tiemann JKS
   20. Trellet M
   21. Woods C
   22. Zhmurov A

   (2019) Sharing data from molecular simulations

   Journal of Chemical Information and Modeling 59:4093–4099.

   https://doi.org/10.1021/acs.jcim.9b00665

   • PubMed
   • Google Scholar
3. 1. Abriata LA
   2. Lepore R
   3. Dal Peraro M

   (2020) About the need to make computational models of biological macromolecules available and discoverable

   Bioinformatics 36:2952–2954.

   https://doi.org/10.1093/bioinformatics/btaa086

   • PubMed
   • Google Scholar
4. 1. Aldeghi M
   2. Heifetz A
   3. Bodkin MJ
   4. Knapp S
   5. Biggin PC

   (2016) Accurate calculation of the absolute free energy of binding for drug molecules

   Chemical Science 7:207–218.

   https://doi.org/10.1039/c5sc02678d

   • PubMed
   • Google Scholar
5. 1. Alessandri R
   2. Sami S
   3. Barnoud J
   4. de Vries AH
   5. Marrink SJ
   6. Havenith RWA

   (2020) Resolving donor–acceptor interfaces and charge carrier energy levels of organic semiconductors with polar side chains

   Advanced Functional Materials 30:2004799.

   https://doi.org/10.1002/adfm.202004799

   • Google Scholar
6. 1. Alessandri R
   2. Grünewald F
   3. Marrink SJ

   (2021) The martini model in materials science

   Advanced Materials 33:e2008635.

   https://doi.org/10.1002/adma.202008635

   • PubMed
   • Google Scholar
7. 1. Amaro RE
   2. Mulholland AJ

   (2020) A community letter regarding sharing biomolecular simulation data for COVID-19

   Journal of Chemical Information and Modeling 60:2653–2656.

   https://doi.org/10.1021/acs.jcim.0c00319

   • PubMed
   • Google Scholar
8. 1. Antila HS
   2. M Ferreira T
   3. Ollila OHS
   4. Miettinen MS

   (2021) Using open data to rapidly benchmark biomolecular simulations: Phospholipid conformational dynamics

   Journal of Chemical Information and Modeling 61:938–949.

   https://doi.org/10.1021/acs.jcim.0c01299

   • PubMed
   • Google Scholar
9. 1. Armstrong DR
   2. Berrisford JM
   3. Conroy MJ
   4. Gutmanas A
   5. Anyango S
   6. Choudhary P
   7. Clark AR
   8. Dana JM
   9. Deshpande M
   10. Dunlop R
   11. Gane P
   12. Gáborová R
   13. Gupta D
   14. Haslam P
   15. Koča J
   16. Mak L
   17. Mir S
   18. Mukhopadhyay A
   19. Nadzirin N
   20. Nair S
   21. Paysan-Lafosse T
   22. Pravda L
   23. Sehnal D
   24. Salih O
   25. Smart O
   26. Tolchard J
   27. Varadi M
   28. Svobodova-Vařeková R
   29. Zaki H
   30. Kleywegt GJ
   31. Velankar S

   (2020) PDBe: improved findability of macromolecular structure data in the PDB

   Nucleic Acids Research 48:D335–D343.

   https://doi.org/10.1093/nar/gkz990

   • PubMed
   • Google Scholar
10. Software

    1. Auer S

    (2018) Towards an open research knowledge graph, version Version 1

    Zenodo.

    https://doi.org/10.5281/zenodo.1157185
11. 1. Berendsen HJC
    2. Postma JPM
    3. van Gunsteren WF
    4. DiNola A
    5. Haak JR

    (1984) Molecular dynamics with coupling to an external bath

    The Journal of Chemical Physics 81:3684–3690.

    https://doi.org/10.1063/1.448118

    • Google Scholar
12. 1. Berendsen HJC
    2. van der Spoel D
    3. van Drunen R

    (1995) GROMACS: A message-passing parallel molecular dynamics implementation

    Computer Physics Communications 91:43–56.

    https://doi.org/10.1016/0010-4655(95)00042-E

    • Google Scholar
13. 1. Berman HM
    2. Westbrook J
    3. Feng Z
    4. Gilliland G
    5. Bhat TN
    6. Weissig H
    7. Shindyalov IN
    8. Bourne PE

    (2000) The protein data bank

    Nucleic Acids Research 28:235–242.

    https://doi.org/10.1093/nar/28.1.235

    • PubMed
    • Google Scholar
14. 1. Berman H
    2. Henrick K
    3. Nakamura H

    (2003) Announcing the worldwide protein data bank

    Nature Structural Biology 10:980.

    https://doi.org/10.1038/nsb1203-980

    • PubMed
    • Google Scholar
15. 1. Bernetti M
    2. Bussi G

    (2020) Pressure control using stochastic cell rescaling

    The Journal of Chemical Physics 153:114107.

    https://doi.org/10.1063/5.0020514

    • PubMed
    • Google Scholar
16. 1. Bonomi M
    2. Bussi G
    3. Camilloni C
    4. Tribello GA
    5. Banáš P
    6. Barducci A
    7. Bernetti M

    (2019) Promoting transparency and reproducibility in enhanced molecular simulations

    Nature Methods 16:670–673.

    https://doi.org/10.1038/s41592-019-0506-8

    • Google Scholar
17. 1. Bottaro S
    2. Lindorff-Larsen K

    (2018) Biophysical experiments and biomolecular simulations: A perfect match?

    Science 361:355–360.

    https://doi.org/10.1126/science.aat4010

    • PubMed
    • Google Scholar
18. Conference

    1. Bowers KJ
    2. Chow DE
    3. Xu H
    4. Dror RO
    5. Eastwood MP
    6. Gregersen BA
    7. Klepeis JL
    8. Kolossvary I
    9. Moraes MA
    10. Sacerdoti FD
    11. Salmon JK
    12. Shan Y
    13. Shaw DE

    (2006) Scalable Algorithms for Molecular Dynamics Simulations on Commodity Clusters

    SC 2006 Proceedings Supercomputing.

    https://doi.org/10.1109/SC.2006.54

    • Google Scholar
19. Preprint

    1. Bran AM
    2. Cox S
    3. White AD
    4. Schwaller P

    (2023) ChemCrow: augmenting large-language models with chemistry tools

    arXiv.

    https://arxiv.org/abs/2304.05376

    • Google Scholar
20. 1. Brooks BR
    2. Brooks CL
    3. Mackerell AD
    4. Nilsson L
    5. Petrella RJ
    6. Roux B
    7. Won Y
    8. Archontis G
    9. Bartels C
    10. Boresch S
    11. Caflisch A
    12. Caves L
    13. Cui Q
    14. Dinner AR
    15. Feig M
    16. Fischer S
    17. Gao J
    18. Hodoscek M
    19. Im W
    20. Kuczera K
    21. Lazaridis T
    22. Ma J
    23. Ovchinnikov V
    24. Paci E
    25. Pastor RW
    26. Post CB
    27. Pu JZ
    28. Schaefer M
    29. Tidor B
    30. Venable RM
    31. Woodcock HL
    32. Wu X
    33. Yang W
    34. York DM
    35. Karplus M

    (2009) CHARMM: the biomolecular simulation program

    Journal of Computational Chemistry 30:1545–1614.

    https://doi.org/10.1002/jcc.21287

    • PubMed
    • Google Scholar
21. 1. Burley SK
    2. Kurisu G
    3. Markley JL
    4. Nakamura H
    5. Velankar S
    6. Berman HM
    7. Sali A
    8. Schwede T
    9. Trewhella J

    (2017) PDB-Dev: A prototype system for depositing integrative/hybrid structural models

    Structure 25:1317–1318.

    https://doi.org/10.1016/j.str.2017.08.001

    • PubMed
    • Google Scholar
22. 1. Bussi G
    2. Donadio D
    3. Parrinello M

    (2007) Canonical sampling through velocity rescaling

    The Journal of Chemical Physics 126:014101.

    https://doi.org/10.1063/1.2408420

    • PubMed
    • Google Scholar
23. 1. Commun Biol

    (2023) Reliability and reproducibility checklist for molecular dynamics simulations

    Communications Biology 6:268.

    https://doi.org/10.1038/s42003-023-04653-0

    • PubMed
    • Google Scholar
24. 1. Dandekar BR
    2. Mondal J

    (2020) Capturing protein-ligand recognition pathways in coarse-grained simulation

    The Journal of Physical Chemistry Letters 11:5302–5311.

    https://doi.org/10.1021/acs.jpclett.0c01683

    • PubMed
    • Google Scholar
25. Conference

    1. Di Cosmo R
    2. Zacchiroli S

    (2017)

    Software Heritage: Why and How to Preserve Software Source Code

    iPRES 2017 - 14th International Conference on Digital Preservation. pp. 1–10.

    • Google Scholar
26. 1. Domański J
    2. Stansfeld PJ
    3. Sansom MSP
    4. Beckstein O

    (2010) Lipidbook: a public repository for force-field parameters used in membrane simulations

    The Journal of Membrane Biology 236:255–258.

    https://doi.org/10.1007/s00232-010-9296-8

    • PubMed
    • Google Scholar
27. 1. Duncan AL
    2. Corey RA
    3. Sansom MSP

    (2020) Defining how multiple lipid species interact with inward rectifier potassium (Kir2) channels

    PNAS 117:7803–7813.

    https://doi.org/10.1073/pnas.1918387117

    • PubMed
    • Google Scholar
28. 1. Dybeck EC
    2. Thiel A
    3. Schnieders MJ
    4. Pickard FC
    5. Wood GPF
    6. Krzyzaniak JF
    7. Hancock BC

    (2023) A comparison of methods for computing relative anhydrous–hydrate stability with molecular simulation

    Crystal Growth & Design 23:142–167.

    https://doi.org/10.1021/acs.cgd.2c00832

    • Google Scholar
29. 1. Elofsson A
    2. Hess B
    3. Lindahl E
    4. Onufriev A
    5. van der Spoel D
    6. Wallqvist A

    (2019) Ten simple rules on how to create open access and reproducible molecular simulations of biological systems

    PLOS Computational Biology 15:e1006649.

    https://doi.org/10.1371/journal.pcbi.1006649

    • PubMed
    • Google Scholar
30. Software

    1. European Organization For Nuclear Research

    (2013) Zenodo

    OpenAIRE.

    https://catalogue.openaire.eu/service/openaire.zenodo/overview
31. 1. Fadda E

    (2022) Molecular simulations of complex carbohydrates and glycoconjugates

    Current Opinion in Chemical Biology 69:102175.

    https://doi.org/10.1016/j.cbpa.2022.102175

    • PubMed
    • Google Scholar
32. 1. Fan FJ
    2. Shi Y

    (2022) Effects of data quality and quantity on deep learning for protein-ligand binding affinity prediction

    Bioorganic & Medicinal Chemistry 72:117003.

    https://doi.org/10.1016/j.bmc.2022.117003

    • Google Scholar
33. 1. Färber M
    2. Lamprecht D

    (2021) The data set knowledge graph: Creating a linked open data source for data sets

    Quantitative Science Studies 2:1324–1355.

    https://doi.org/10.1162/qss_a_00161

    • Google Scholar
34. 1. Fawzi NL
    2. Parekh SH
    3. Mittal J

    (2021) Biophysical studies of phase separation integrating experimental and computational methods

    Current Opinion in Structural Biology 70:78–86.

    https://doi.org/10.1016/j.sbi.2021.04.004

    • PubMed
    • Google Scholar
35. 1. Fuller JC
    2. Jackson RM
    3. Edwards TA
    4. Wilson AJ
    5. Shirts MR

    (2012) Modeling of arylamide helix mimetics in the P53 peptide binding site of hDM2 suggests parallel and anti-parallel conformations are both stable

    PLOS ONE 7:e43253.

    https://doi.org/10.1371/journal.pone.0043253

    • PubMed
    • Google Scholar
36. 1. Genheden S
    2. Ryde U

    (2015) The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities

    Expert Opinion on Drug Discovery 10:449–461.

    https://doi.org/10.1517/17460441.2015.1032936

    • PubMed
    • Google Scholar
37. 1. Gertsen AS
    2. Sørensen MK
    3. Andreasen JW

    (2020) Nanostructure of organic semiconductor thin films: Molecular dynamics modeling with solvent evaporation

    Physical Review Materials 4:075405.

    https://doi.org/10.1103/PhysRevMaterials.4.075405

    • Google Scholar
38. Conference

    1. Gowers R
    2. Linke M
    3. Barnoud J
    4. Reddy T
    5. Melo M
    6. Seyler S
    7. Domański J
    8. Dotson D
    9. Buchoux S
    10. Kenney I
    11. Beckstein O

    (2016) MDAnalysis: A Python Package for the Rapid Analysis of Molecular Dynamics Simulations

    Python in Science Conference.

    https://doi.org/10.25080/Majora-629e541a-00e

    • Google Scholar
39. 1. Gupta C
    2. Sarkar D
    3. Tieleman DP
    4. Singharoy A

    (2022) The ugly, bad, and good stories of large-scale biomolecular simulations

    Current Opinion in Structural Biology 73:102338.

    https://doi.org/10.1016/j.sbi.2022.102338

    • PubMed
    • Google Scholar
40. 1. Hénin J
    2. Lelièvre T
    3. Shirts MR
    4. Valsson O
    5. Delemotte L

    (2022) Enhanced sampling methods for molecular dynamics simulations [Article v1.0]

    Living Journal of Computational Molecular Science 4:1583.

    https://doi.org/10.33011/livecoms.4.1.1583

    • Google Scholar
41. 1. Hoch JC
    2. Baskaran K
    3. Burr H
    4. Chin J
    5. Eghbalnia HR
    6. Fujiwara T
    7. Gryk MR
    8. Iwata T
    9. Kojima C
    10. Kurisu G
    11. Maziuk D
    12. Miyanoiri Y
    13. Wedell JR
    14. Wilburn C
    15. Yao H
    16. Yokochi M

    (2023) Biological magnetic resonance data bank

    Nucleic Acids Research 51:D368–D376.

    https://doi.org/10.1093/nar/gkac1050

    • PubMed
    • Google Scholar
42. 1. Hollingsworth SA
    2. Dror RO

    (2018) Molecular dynamics simulation for all

    Neuron 99:1129–1143.

    https://doi.org/10.1016/j.neuron.2018.08.011

    • PubMed
    • Google Scholar
43. 1. Hospital A
    2. Battistini F
    3. Soliva R
    4. Gelpí JL
    5. Orozco M

    (2020) Surviving the deluge of biosimulation data

    WIREs Computational Molecular Science 10:e1449.

    https://doi.org/10.1002/wcms.1449

    • Google Scholar
44. 1. Humphrey W
    2. Dalke A
    3. Schulten K

    (1996) VMD: visual molecular dynamics

    Journal of Molecular Graphics 14:33–38.

    https://doi.org/10.1016/0263-7855(96)00018-5

    • PubMed
    • Google Scholar
45. 1. Ingólfsson HI
    2. Lopez CA
    3. Uusitalo JJ
    4. de Jong DH
    5. Gopal SM
    6. Periole X
    7. Marrink SJ

    (2014) The power of coarse graining in biomolecular simulations

    Wiley Interdisciplinary Reviews. Computational Molecular Science 4:225–248.

    https://doi.org/10.1002/wcms.1169

    • PubMed
    • Google Scholar
46. 1. Ivanov P
    2. Mu J
    3. Leay L
    4. Chang SY
    5. Sharrad CA
    6. Masters AJ
    7. Schroeder SLM

    (2017) Organic and Third Phase in HNO3/TBP/n-Dodecane System: No Reverse Micelles

    Solvent Extraction and Ion Exchange 35:251–265.

    https://doi.org/10.1080/07366299.2017.1336048

    • Google Scholar
47. 1. Jahani F
    2. Torabi S
    3. Chiechi RC
    4. Koster LJA
    5. Hummelen JC

    (2014) Fullerene derivatives with increased dielectric constants

    Chemical Communications 50:10645–10647.

    https://doi.org/10.1039/c4cc04366a

    • PubMed
    • Google Scholar
48. 1. Jeon JH
    2. Javanainen M
    3. Martinez-Seara H
    4. Metzler R
    5. Vattulainen I

    (2016) Protein crowding in lipid bilayers gives rise to non-gaussian anomalous lateral diffusion of phospholipids and proteins

    Physical Review X 6:021006.

    https://doi.org/10.1103/PhysRevX.6.021006

    • Google Scholar
49. Data

    1. Johns Hopkins University

    (authors) (2020) COVID-19 data repository by the center for systems science and engineering (CSSE) at johns hopkins university

    GitHub.

    https://github.com/CSSEGISandData/COVID-19
50. 1. Jumper J
    2. Evans R
    3. Pritzel A
    4. Green T
    5. Figurnov M
    6. Ronneberger O
    7. Tunyasuvunakool K
    8. Bates R
    9. Žídek A
    10. Potapenko A
    11. Bridgland A
    12. Meyer C
    13. Kohl SAA
    14. Ballard AJ
    15. Cowie A
    16. Romera-Paredes B
    17. Nikolov S
    18. Jain R
    19. Adler J
    20. Back T
    21. Petersen S
    22. Reiman D
    23. Clancy E
    24. Zielinski M
    25. Steinegger M
    26. Pacholska M
    27. Berghammer T
    28. Bodenstein S
    29. Silver D
    30. Vinyals O
    31. Senior AW
    32. Kavukcuoglu K
    33. Kohli P
    34. Hassabis D

    (2021) Highly accurate protein structure prediction with AlphaFold

    Nature 596:583–589.

    https://doi.org/10.1038/s41586-021-03819-2

    • PubMed
    • Google Scholar
51. 1. Kabelka I
    2. Brožek R
    3. Vácha R

    (2021) Selecting collective variables and free-energy methods for peptide translocation across membranes

    Journal of Chemical Information and Modeling 61:819–830.

    https://doi.org/10.1021/acs.jcim.0c01312

    • PubMed
    • Google Scholar
52. 1. Kampfrath M
    2. Staritzbichler R
    3. Hernández GP
    4. Rose AS
    5. Tiemann JKS
    6. Scheuermann G
    7. Wiegreffe D
    8. Hildebrand PW

    (2022) MDsrv: visual sharing and analysis of molecular dynamics simulations

    Nucleic Acids Research 50:W483–W489.

    https://doi.org/10.1093/nar/gkac398

    • PubMed
    • Google Scholar
53. 1. Karunasena C
    2. Li S
    3. Heifner MC
    4. Ryno SM
    5. Risko C

    (2021) Reconsidering the roles of noncovalent intramolecular “locks” in π-conjugated molecules

    Chemistry of Materials 33:9139–9151.

    https://doi.org/10.1021/acs.chemmater.1c02335

    • Google Scholar
54. 1. Kelly BD
    2. Smith WR

    (2020) Alchemical hydration free-energy calculations using molecular dynamics with explicit polarization and induced polarity decoupling: An on-the-fly polarization approach

    Journal of Chemical Theory and Computation 16:1146–1161.

    https://doi.org/10.1021/acs.jctc.9b01139

    • PubMed
    • Google Scholar
55. Preprint

    1. Kiirikki A
    2. Antila H
    3. Bort L
    4. Buslaev P
    5. Fernando F
    6. Mendes Ferreira T
    7. Fuchs P
    8. Garcia-Fandino R
    9. Gushchin I
    10. Kav B
    11. Kučerka N
    12. Kula P
    13. Kurki M
    14. Kuzmin A
    15. Madsen J
    16. Miettinen M
    17. Nencini R
    18. Piggot T
    19. Pineiro A
    20. Suarez-Leston F
    21. Ollila S

    (2023) NMRlipids Databank Makes Data-Driven Analysis of Biomembrane Properties Accessible for All

    ChemRxiv.

    https://doi.org/10.26434/chemrxiv-2023-jrpwm-v2

    • Google Scholar
56. 1. Kinjo AR
    2. Bekker GJ
    3. Suzuki H
    4. Tsuchiya Y
    5. Kawabata T
    6. Ikegawa Y
    7. Nakamura H

    (2017) Protein Data Bank Japan (PDBj): updated user interfaces, resource description framework, analysis tools for large structures

    Nucleic Acids Research 45:D282–D288.

    https://doi.org/10.1093/nar/gkw962

    • PubMed
    • Google Scholar
57. 1. Kirschner KN
    2. Yongye AB
    3. Tschampel SM
    4. González-Outeiriño J
    5. Daniels CR
    6. Foley BL
    7. Woods RJ

    (2008) GLYCAM06: a generalizable biomolecular force field. Carbohydrates

    Journal of Computational Chemistry 29:622–655.

    https://doi.org/10.1002/jcc.20820

    • PubMed
    • Google Scholar
58. 1. Krishna S
    2. Sreedhar I
    3. Patel CM

    (2021) Molecular dynamics simulation of polyamide-based materials – A review

    Computational Materials Science 200:110853.

    https://doi.org/10.1016/j.commatsci.2021.110853

    • Google Scholar
59. 1. Kümmerer F
    2. Orioli S
    3. Harding-Larsen D
    4. Hoffmann F
    5. Gavrilov Y
    6. Teilum K
    7. Lindorff-Larsen K

    (2021) Fitting side-chain nmr relaxation data using molecular simulations

    Journal of Chemical Theory and Computation 17:5262–5275.

    https://doi.org/10.1021/acs.jctc.0c01338

    • PubMed
    • Google Scholar
60. 1. Kyrychenko A
    2. Karpushina GV
    3. Svechkarev D
    4. Kolodezny D
    5. Bogatyrenko SI
    6. Kryshtal AP
    7. Doroshenko AO

    (2012) Fluorescence probing of thiol-functionalized gold nanoparticles: Is alkylthiol coating of a nanoparticle as hydrophobic as expected?

    The Journal of Physical Chemistry C 116:21059–21068.

    https://doi.org/10.1021/jp3060813

    • Google Scholar
61. 1. Lane TJ

    (2023) Protein structure prediction has reached the single-structure frontier

    Nature Methods 20:170–173.

    https://doi.org/10.1038/s41592-022-01760-4

    • PubMed
    • Google Scholar
62. 1. Liu S
    2. Cao S
    3. Hoang K
    4. Young KL
    5. Paluch AS
    6. Mobley DL

    (2016) Using MD simulations to calculate how solvents modulate solubility

    Journal of Chemical Theory and Computation 12:1930–1941.

    https://doi.org/10.1021/acs.jctc.5b00934

    • PubMed
    • Google Scholar
63. 1. Mahmud M
    2. Kaiser MS
    3. McGinnity TM
    4. Hussain A

    (2021) Deep learning in mining biological data

    Cognitive Computation 13:1–33.

    https://doi.org/10.1007/s12559-020-09773-x

    • PubMed
    • Google Scholar
64. 1. Marklund EG
    2. Benesch JL

    (2019) Weighing-up protein dynamics: the combination of native mass spectrometry and molecular dynamics simulations

    Current Opinion in Structural Biology 54:50–58.

    https://doi.org/10.1016/j.sbi.2018.12.011

    • Google Scholar
65. 1. Martinez X
    2. Baaden M

    (2021) UnityMol prototype for FAIR sharing of molecular-visualization experiences: from pictures in the cloud to collaborative virtual reality exploration in immersive 3D environments

    Acta Crystallographica. Section D, Structural Biology 77:746–754.

    https://doi.org/10.1107/S2059798321002941

    • PubMed
    • Google Scholar
66. 1. Marx V

    (2013) Biology: The big challenges of big data

    Nature 498:255–260.

    https://doi.org/10.1038/498255a

    • PubMed
    • Google Scholar
67. Conference

    1. McKinney W

    (2010) Data Structures for Statistical Computing in Python

    Python in Science Conference. pp. 56–61.

    https://doi.org/10.25080/Majora-92bf1922-00a

    • Google Scholar
68. 1. Merz KM
    2. Amaro R
    3. Cournia Z
    4. Rarey M
    5. Soares T
    6. Tropsha A
    7. Wahab HA
    8. Wang R

    (2020) Editorial: Method and data sharing and reproducibility of scientific results

    Journal of Chemical Information and Modeling 60:5868–5869.

    https://doi.org/10.1021/acs.jcim.0c01389

    • PubMed
    • Google Scholar
69. 1. Meyer T
    2. D’Abramo M
    3. Hospital A
    4. Rueda M
    5. Ferrer-Costa C
    6. Pérez A
    7. Carrillo O
    8. Camps J
    9. Fenollosa C
    10. Repchevsky D
    11. Gelpí JL
    12. Orozco M

    (2010) MoDEL (Molecular Dynamics Extended Library): a database of atomistic molecular dynamics trajectories

    Structure 18:1399–1409.

    https://doi.org/10.1016/j.str.2010.07.013

    • PubMed
    • Google Scholar
70. 1. Michaud-Agrawal N
    2. Denning EJ
    3. Woolf TB
    4. Beckstein O

    (2011) MDAnalysis: A toolkit for the analysis of molecular dynamics simulations

    Journal of Computational Chemistry 32:2319–2327.

    https://doi.org/10.1002/jcc.21787

    • PubMed
    • Google Scholar
71. 1. Miksa T
    2. Simms S
    3. Mietchen D
    4. Jones S

    (2019) Ten principles for machine-actionable data management plans

    PLOS Computational Biology 15:e1006750.

    https://doi.org/10.1371/journal.pcbi.1006750

    • PubMed
    • Google Scholar
72. 1. Mulholland AJ
    2. Amaro RE

    (2020) COVID19 - Computational Chemists Meet the Moment

    Journal of Chemical Information and Modeling 60:5724–5726.

    https://doi.org/10.1021/acs.jcim.0c01395

    • PubMed
    • Google Scholar
73. 1. Musen MA

    (2022) Without appropriate metadata, data-sharing mandates are pointless

    Nature 609:222.

    https://doi.org/10.1038/d41586-022-02820-7

    • PubMed
    • Google Scholar
74. 1. Newport TD
    2. Sansom MSP
    3. Stansfeld PJ

    (2019) The MemProtMD database: a resource for membrane-embedded protein structures and their lipid interactions

    Nucleic Acids Research 47:D390–D397.

    https://doi.org/10.1093/nar/gky1047

    • Google Scholar
75. 1. Oliphant TE

    (2007) Python for scientific computing

    Computing in Science & Engineering 9:10–20.

    https://doi.org/10.1109/MCSE.2007.58

    • Google Scholar
76. Software

    1. Panero P
    2. Benito J

    (2022) OpenAIRE webinar: Zenodo - open digital repository, version Version v1

    Zenodo.

    https://doi.org/10.5281/zenodo.7417839
77. 1. Parrinello M
    2. Rahman A

    (1981) Polymorphic transitions in single crystals: A new molecular dynamics method

    Journal of Applied Physics 52:7182–7190.

    https://doi.org/10.1063/1.328693

    • Google Scholar
78. 1. Perera N
    2. Dehmer M
    3. Emmert-Streib F

    (2020) Named entity recognition and relation detection for biomedical information extraction

    Frontiers in Cell and Developmental Biology 8:673.

    https://doi.org/10.3389/fcell.2020.00673

    • PubMed
    • Google Scholar
79. 1. Perilla JR
    2. Goh BC
    3. Cassidy CK
    4. Liu B
    5. Bernardi RC
    6. Rudack T
    7. Yu H
    8. Wu Z
    9. Schulten K

    (2015) Molecular dynamics simulations of large macromolecular complexes

    Current Opinion in Structural Biology 31:64–74.

    https://doi.org/10.1016/j.sbi.2015.03.007

    • PubMed
    • Google Scholar
80. 1. Perkel J

    (2016) Democratic databases: science on GitHub

    Nature 538:127–128.

    https://doi.org/10.1038/538127a

    • PubMed
    • Google Scholar
81. Preprint

    1. Pesce F
    2. Lindorff-Larsen K

    (2023) Combining Experiments and Simulations to Examine the Temperature-Dependent Behaviour of a Disordered Protein

    bioRxiv.

    https://doi.org/10.1101/2023.03.04.531094

    • Google Scholar
82. 1. Pettersen EF
    2. Goddard TD
    3. Huang CC
    4. Couch GS
    5. Greenblatt DM
    6. Meng EC
    7. Ferrin TE

    (2004) UCSF Chimera--A visualization system for exploratory research and analysis

    Journal of Computational Chemistry 25:1605–1612.

    https://doi.org/10.1002/jcc.20084

    • PubMed
    • Google Scholar
83. 1. Phillips JC
    2. Hardy DJ
    3. Maia JDC
    4. Stone JE
    5. Ribeiro JV
    6. Bernardi RC
    7. Buch R
    8. Fiorin G
    9. Hénin J
    10. Jiang W
    11. McGreevy R
    12. Melo MCR
    13. Radak BK
    14. Skeel RD
    15. Singharoy A
    16. Wang Y
    17. Roux B
    18. Aksimentiev A
    19. Luthey-Schulten Z
    20. Kalé LV
    21. Schulten K
    22. Chipot C
    23. Tajkhorshid E

    (2020) Scalable molecular dynamics on CPU and GPU architectures with NAMD

    The Journal of Chemical Physics 153:044130.

    https://doi.org/10.1063/5.0014475

    • PubMed
    • Google Scholar
84. 1. Piskorz TK
    2. Gobbo C
    3. Marrink SJ
    4. De Feyter S
    5. de Vries AH
    6. van Esch JH

    (2019) Nucleation mechanisms of self-assembled physisorbed monolayers on graphite

    The Journal of Physical Chemistry C 123:17510–17520.

    https://doi.org/10.1021/acs.jpcc.9b01234

    • Google Scholar
85. 1. Pohjolainen E
    2. Chen X
    3. Malola S
    4. Groenhof G
    5. Häkkinen H

    (2016) A Unified AMBER-compatible molecular mechanics force field for thiolate-protected gold nanoclusters

    Journal of Chemical Theory and Computation 12:1342–1350.

    https://doi.org/10.1021/acs.jctc.5b01053

    • PubMed
    • Google Scholar
86. 1. Porubsky VL
    2. Goldberg AP
    3. Rampadarath AK
    4. Nickerson DP
    5. Karr JR
    6. Sauro HM

    (2020) Best practices for making reproducible biochemical models

    Cell Systems 11:109–120.

    https://doi.org/10.1016/j.cels.2020.06.012

    • PubMed
    • Google Scholar
87. Software

    1. Poulain P

    (2023) MDverse data analysis, version swh:1:rev:4562c50d1b51a51fdf952ae6e9efaa407dd06e20

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:fc72ac7a9c9f0489a361cb2b7fcf8ba48898e4ee;origin=https://github.com/MDverse/mdda;visit=swh:1:snp:dbfe8b4401ac98d3728ebb00241429274c619beb;anchor=swh:1:rev:4562c50d1b51a51fdf952ae6e9efaa407dd06e20
88. Software

    1. Poulain P
    2. Bouarroudj L
    3. Tiemann JKS
    4. Bussi G

    (2023) MDverse web scrapper, version swh:1:rev:0524199041e84be2d69993540ad8e2223d3b4698

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:ce91602834cf79e634d26aff585a9fea22b0fea3;origin=https://github.com/MDverse/mdws;visit=swh:1:snp:540580756b211c116bd602423e0262d3055b8251;anchor=swh:1:rev:0524199041e84be2d69993540ad8e2223d3b4698
89. Software

    1. Poulain P
    2. Oussaren M

    (2023) MDverse data explorer, version swh:1:rev:52604906f80f96b27fd61209a78a93cd36be9a45

    Software Heritage.

    https://archive.softwareheritage.org/swh:1:dir:1fc8b8eaabf4a9087e6d5b0ec5ed97031482bcbf;origin=https://github.com/MDverse/mdde;visit=swh:1:snp:5a3326fd135f604290fb799470f524384a959b04;anchor=swh:1:rev:52604906f80f96b27fd61209a78a93cd36be9a45
90. 1. Qiu L
    2. Liu J
    3. Alessandri R
    4. Qiu X
    5. Koopmans M
    6. Havenith RWA
    7. Marrink SJ
    8. Chiechi RC
    9. Anton Koster LJ
    10. Hummelen JC

    (2017) Enhancing doping efficiency by improving host-dopant miscibility for fullerene-based n-type thermoelectrics

    Journal of Materials Chemistry A 5:21234–21241.

    https://doi.org/10.1039/C7TA06609K

    • Google Scholar
91. 1. Rodríguez-Espigares I
    2. Torrens-Fontanals M
    3. Tiemann JKS
    4. Aranda-García D
    5. Ramírez-Anguita JM
    6. Stepniewski TM
    7. Worp N
    8. Varela-Rial A
    9. Morales-Pastor A
    10. Medel-Lacruz B
    11. Pándy-Szekeres G
    12. Mayol E
    13. Giorgino T
    14. Carlsson J
    15. Deupi X
    16. Filipek S
    17. Filizola M
    18. Gómez-Tamayo JC
    19. Gonzalez A
    20. Gutiérrez-de-Terán H
    21. Jiménez-Rosés M
    22. Jespers W
    23. Kapla J
    24. Khelashvili G
    25. Kolb P
    26. Latek D
    27. Marti-Solano M
    28. Matricon P
    29. Matsoukas M-T
    30. Miszta P
    31. Olivella M
    32. Perez-Benito L
    33. Provasi D
    34. Ríos S
    35. R Torrecillas I
    36. Sallander J
    37. Sztyler A
    38. Vasile S
    39. Weinstein H
    40. Zachariae U
    41. Hildebrand PW
    42. De Fabritiis G
    43. Sanz F
    44. Gloriam DE
    45. Cordomi A
    46. Guixà-González R
    47. Selent J

    (2020) GPCRmd uncovers the dynamics of the 3D-GPCRome

    Nature Methods 17:777–787.

    https://doi.org/10.1038/s41592-020-0884-y

    • PubMed
    • Google Scholar
92. 1. Salomon‐Ferrer R
    2. Case DA
    3. Walker RC

    (2013) An overview of the Amber biomolecular simulation package

    WIREs Computational Molecular Science 3:198–210.

    https://doi.org/10.1002/wcms.1121

    • Google Scholar
93. 1. Sami S
    2. Alessandri R
    3. W. Wijaya JB
    4. Grünewald F
    5. de Vries AH
    6. Marrink SJ
    7. Broer R
    8. Havenith RWA

    (2022) Strategies for enhancing the dielectric constant of organic materials

    The Journal of Physical Chemistry C 126:19462–19469.

    https://doi.org/10.1021/acs.jpcc.2c05682

    • Google Scholar
94. 1. Sarkar A
    2. Sasmal R
    3. Empereur-mot C
    4. Bochicchio D
    5. Kompella SVK
    6. Sharma K
    7. Dhiman S
    8. Sundaram B
    9. Agasti SS
    10. Pavan GM
    11. George SJ

    (2020) Self-sorted, random, and block supramolecular copolymers via sequence controlled, multicomponent self-assembly

    Journal of the American Chemical Society 142:7606–7617.

    https://doi.org/10.1021/jacs.0c01822

    • Google Scholar
95. 1. Schaefer SL
    2. Hummer G

    (2022) Sublytic gasdermin-D pores captured in atomistic molecular simulations

    eLife 11:e81432.

    https://doi.org/10.7554/eLife.81432

    • PubMed
    • Google Scholar
96. 1. Souza PCT
    2. Alessandri R
    3. Barnoud J
    4. Thallmair S
    5. Faustino I
    6. Grünewald F
    7. Patmanidis I
    8. Abdizadeh H
    9. Bruininks BMH
    10. Wassenaar TA
    11. Kroon PC
    12. Melcr J
    13. Nieto V
    14. Corradi V
    15. Khan HM
    16. Domański J
    17. Javanainen M
    18. Martinez-Seara H
    19. Reuter N
    20. Best RB
    21. Vattulainen I
    22. Monticelli L
    23. Periole X
    24. Tieleman DP
    25. de Vries AH
    26. Marrink SJ

    (2021) Martini 3: a general purpose force field for coarse-grained molecular dynamics

    Nature Methods 18:382–388.

    https://doi.org/10.1038/s41592-021-01098-3

    • PubMed
    • Google Scholar
97. 1. Stansfeld PJ
    2. Goose JE
    3. Caffrey M
    4. Carpenter EP
    5. Parker JL
    6. Newstead S
    7. Sansom MSP

    (2015) MemProtMD: Automated insertion of membrane protein structures into explicit lipid membranes

    Structure 23:1350–1361.

    https://doi.org/10.1016/j.str.2015.05.006

    • PubMed
    • Google Scholar
98. 1. Stephens ZD
    2. Lee SY
    3. Faghri F
    4. Campbell RH
    5. Zhai C
    6. Efron MJ
    7. Iyer R
    8. Schatz MC
    9. Sinha S
    10. Robinson GE

    (2015) Big data: Astronomical or genomical?

    PLOS Biology 13:e1002195.

    https://doi.org/10.1371/journal.pbio.1002195

    • PubMed
    • Google Scholar
99. 1. Sud M
    2. Fahy E
    3. Cotter D
    4. Brown A
    5. Dennis EA
    6. Glass CK
    7. Merrill AH Jr
    8. Murphy RC
    9. Raetz CRH
    10. Russell DW
    11. Subramaniam S

    (2007) LMSD: LIPID MAPS structure database

    Nucleic Acids Research 35:D527–D532.

    https://doi.org/10.1093/nar/gkl838

    • PubMed
    • Google Scholar
100. 1. Tai K
     2. Murdock S
     3. Wu B
     4. Ng MH
     5. Johnston S
     6. Fangohr H
     7. Cox SJ
     8. Jeffreys P
     9. Essex JW
     10. P. Sansom MS

     (2004) BioSimGrid: towards a worldwide repository for biomolecular simulations

     Organic &Biomolecular Chemistry 2:3219.

     https://doi.org/10.1039/b411352g

     • Google Scholar
101. 1. Tiemann JKS
     2. Guixà-González R
     3. Hildebrand PW
     4. Rose AS

     (2017) MDsrv: viewing and sharing molecular dynamics simulations on the web

     Nature Methods 14:1123–1124.

     https://doi.org/10.1038/nmeth.4497

     • PubMed
     • Google Scholar
102. 1. UniProt Consortium

     (2022) UniProt: The universal protein knowledgebase in 2023

     Nucleic Acids Research 51:D523–D531.

     https://doi.org/10.1093/nar/gkac1052

     • Google Scholar
103. 1. Vanommeslaeghe K
     2. MacKerell AD

     (2012) Automation of the CHARMM General Force Field (CGenFF) I: bond perception and atom typing

     Journal of Chemical Information and Modeling 52:3144–3154.

     https://doi.org/10.1021/ci300363c

     • PubMed
     • Google Scholar
104. Report

     1. van Rossum G

     (1995) Python Tutorial

     Amsterdam, The Netherlands: Centrum voor Wiskunde en Informatica.

     https://ir.cwi.nl/pub/5007

     • Google Scholar
105. 1. Virtanen SI
     2. Kiirikki AM
     3. Mikula KM
     4. Iwaï H
     5. Ollila OHS

     (2020) Heterogeneous dynamics in partially disordered proteins

     Physical Chemistry Chemical Physics 22:21185–21196.

     https://doi.org/10.1039/d0cp03473h

     • PubMed
     • Google Scholar
106. 1. Vuorio J
     2. Vattulainen I
     3. Martinez-Seara H

     (2017) Atomistic fingerprint of hyaluronan-CD44 binding

     PLOS Computational Biology 13:e1005663.

     https://doi.org/10.1371/journal.pcbi.1005663

     • PubMed
     • Google Scholar
107. 1. Waltemath D
     2. Adams R
     3. Beard DA
     4. Bergmann FT
     5. Bhalla US
     6. Britten R
     7. Chelliah V
     8. Cooling MT
     9. Cooper J
     10. Crampin EJ
     11. Garny A
     12. Hoops S
     13. Hucka M
     14. Hunter P
     15. Klipp E
     16. Laibe C
     17. Miller AK
     18. Moraru I
     19. Nickerson D
     20. Nielsen P
     21. Nikolski M
     22. Sahle S
     23. Sauro HM
     24. Schmidt H
     25. Snoep JL
     26. Tolle D
     27. Wolkenhauer O
     28. Le Novère N

     (2011) Minimum Information About a Simulation Experiment (MIASE)

     PLOS Computational Biology 7:e1001122.

     https://doi.org/10.1371/journal.pcbi.1001122

     • Google Scholar
108. 1. Wang J
     2. Wolf RM
     3. Caldwell JW
     4. Kollman PA
     5. Case DA

     (2004) Development and testing of a general amber force field

     Journal of Computational Chemistry 25:1157–1174.

     https://doi.org/10.1002/jcc.20035

     • PubMed
     • Google Scholar
109. 1. Wilkinson MD
     2. Dumontier M
     3. Aalbersberg IJJ
     4. Appleton G
     5. Axton M
     6. Baak A
     7. Blomberg N
     8. Boiten JW
     9. da Silva Santos LB
     10. Bourne PE
     11. Bouwman J
     12. Brookes AJ
     13. Clark T
     14. Crosas M
     15. Dillo I
     16. Dumon O
     17. Edmunds S
     18. Evelo CT
     19. Finkers R
     20. Gonzalez-Beltran A
     21. Gray AJG
     22. Groth P
     23. Goble C
     24. Grethe JS
     25. Heringa J
     26. ’t Hoen PAC
     27. Hooft R
     28. Kuhn T
     29. Kok R
     30. Kok J
     31. Lusher SJ
     32. Martone ME
     33. Mons A
     34. Packer AL
     35. Persson B
     36. Rocca-Serra P
     37. Roos M
     38. van Schaik R
     39. Sansone SA
     40. Schultes E
     41. Sengstag T
     42. Slater T
     43. Strawn G
     44. Swertz MA
     45. Thompson M
     46. van der Lei J
     47. van Mulligen E
     48. Velterop J
     49. Waagmeester A
     50. Wittenburg P
     51. Wolstencroft K
     52. Zhao J
     53. Mons B

     (2016) The FAIR Guiding Principles for scientific data management and stewardship

     Scientific Data 3:160018.

     https://doi.org/10.1038/sdata.2016.18

     • PubMed
     • Google Scholar
110. 1. Wilson SL
     2. Way GP
     3. Bittremieux W
     4. Armache JP
     5. Haendel MA
     6. Hoffman MM

     (2021) Sharing biological data: why, when, and how

     FEBS Letters 595:847–863.

     https://doi.org/10.1002/1873-3468.14067

     • PubMed
     • Google Scholar
111. 1. Yoo J
     2. Winogradoff D
     3. Aksimentiev A

     (2020) Molecular dynamics simulations of DNA-DNA and DNA-protein interactions

     Current Opinion in Structural Biology 64:88–96.

     https://doi.org/10.1016/j.sbi.2020.06.007

     • PubMed
     • Google Scholar
112. 1. Young TA
     2. Martí-Centelles V
     3. Wang J
     4. Lusby PJ
     5. Duarte F

     (2020) RAtionalizing the activity of an “artificial diels-alderase”: Establishing efficient and accurate protocols for calculating supramolecular catalysis

     Journal of the American Chemical Society 142:1300–1310.

     https://doi.org/10.1021/jacs.9b10302

     • PubMed
     • Google Scholar
113. 1. Zheng X
     2. Chan MHY
     3. Chan AKW
     4. Cao S
     5. Ng M
     6. Sheong FK
     7. Li C
     8. Goonetilleke EC
     9. Lam WWY
     10. Lau TC
     11. Huang X
     12. Yam VWW

     (2022) Elucidation of the key role of Pt···Pt interactions in the directional self-assembly of platinum(II) complexes

     PNAS 119:e2116543119.

     https://doi.org/10.1073/pnas.2116543119

     • PubMed
     • Google Scholar
114. 1. Zhu S

     (2019) Validation of the Generalized Force Fields GAFF, CGenFF, OPLS-AA, and PRODRGFF by Testing Against Experimental Osmotic Coefficient Data for Small Drug-Like Molecules

     Journal of Chemical Information and Modeling 59:4239–4247.

     https://doi.org/10.1021/acs.jcim.9b00552

     • Google Scholar
115. 1. Zhu J
     2. Salvatella X
     3. Robustelli P

     (2022) Small molecules targeting the disordered transactivation domain of the androgen receptor induce the formation of collapsed helical states

     Nature Communications 13:6390.

     https://doi.org/10.1038/s41467-022-34077-z

     • PubMed
     • Google Scholar

Author details

1. Johanna KS Tiemann

   Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark

   Present address

   NovozymesA/S, Lyngby, Denmark

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Supervision, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

   For correspondence

   johanna.tiemann@gmail.com

   Competing interests

   No competing interests declared

2. Magdalena Szczuka

   Institut de Pharmacologie et Biologie Structurale, CNRS, Université de Toulouse, Toulouse, France

   Contribution

   Data curation, Software, Formal analysis, Visualization

   Competing interests

   No competing interests declared

3. Lisa Bouarroudj

   Université Paris Cité, CNRS, Institut Jacques Monod, Paris, France

   Contribution

   Data curation, Software, Formal analysis, Visualization

   Competing interests

   No competing interests declared

4. Mohamed Oussaren

   Université Paris Cité, CNRS, Institut Jacques Monod, Paris, France

   Contribution

   Data curation, Software, Formal analysis, Visualization

   Competing interests

   No competing interests declared

5. Steven Garcia

   Independent researcher, Amsterdam, Netherlands

   Contribution

   Data curation, Software, Formal analysis

   Competing interests

   No competing interests declared

6. Rebecca J Howard

   Department of Biochemistry and Biophysics, Science for Life Laboratory, Stockholm University, Stockholm, Sweden

   Contribution

   Conceptualization, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-2049-3378

7. Lucie Delemotte

   Department of applied physics, Science for Life Laboratory, KTH Royal Institute of Technology, Stockholm, Sweden

   Contribution

   Conceptualization, Writing – original draft, Writing – review and editing

   Competing interests

   Reviewing editor, eLife

8. Erik Lindahl

   1. Department of Biochemistry and Biophysics, Science for Life Laboratory, Stockholm University, Stockholm, Sweden
   2. Department of applied physics, Science for Life Laboratory, KTH Royal Institute of Technology, Stockholm, Sweden

   Contribution

   Conceptualization, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-2734-2794

9. Marc Baaden

   Laboratoire de Biochimie Théorique, CNRS, Université Paris Cité, Paris, France

   Contribution

   Conceptualization, Writing – original draft, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6472-0486

10. Kresten Lindorff-Larsen

    Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark

    Contribution

    Conceptualization, Writing – original draft, Writing – review and editing

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0002-4750-6039

11. Matthieu Chavent

    Institut de Pharmacologie et Biologie Structurale, CNRS, Université de Toulouse, Toulouse, France

    Contribution

    Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

    For correspondence

    matthieu.chavent@ipbs.fr

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-4524-4773

12. Pierre Poulain

    Université Paris Cité, CNRS, Institut Jacques Monod, Paris, France

    Contribution

    Conceptualization, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing – original draft, Writing – review and editing

    For correspondence

    pierre.poulain@u-paris.fr

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-4177-3619

• 1,976

  views

• 158

  downloads

• 7

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.