The art of modeling in the molecular sciences is highly dependent on the available computational technology, underlying data, and ability to collaborate. With the ever-increasing market share of mobile devices, tablets will likely overtake laptops as the computer of choice in the coming years. As the world population shifts to mobile architectures for their primary computing devices, modeling tools need to evolve to provide the required functionality in the portable and interactive tactile environment afforded by these devices. Mobile modeling in the context of molecular sciences is focused on using smartphone or tablet devices to perform a variety of modeling tasks, including but not limited to molecular modeling, geometry optimization, chemical-structure-based calculations, structure-activity relationship (SAR) generation, and database lookups.
The primary benefit of mobile technologies in molecular sciences is the simplistic and intuitive nature of the touchscreen interface that yields a faster learning curve and wider accessibility to nonspecialists. Software is distributed through app stores, which facilitates installation, provides continuous updating, and eliminates the need for centralized information technology (IT) support. The small and portable nature of the devices allows science to be done immediately on idea formation rather than delaying until in an office setting. This makes it ideal for the new generation of telecommuters who can maximize research time and minimize downtime by modeling whenever the opportunity arrives.
Mobile devices also introduce some challenges. The simple touchscreen interface limits the complexity of interactions. For example, a molecular visualization interface (such as Sybyl®, Maestro, or MOE) accessed through a Windows desktop commonly has specific functions based on which mouse button a user presses, such as for atom selection, rotation, and menu call-up. Switching to an interface that has only one type of click and drag action and yet provides equivalent functionality can be difficult. The precision of a finger on a screen is very low compared to that of a mouse or trackpad, and the screen is partially obscured with each operation. In addition, the smaller screen size of portable devices not only makes viewing complex macromolecules difficult, it also limits the real estate available for adding buttons to increase interface functionality.
The benefits and detriments of mobile devices for complicated scientific applications will likely have little effect on the penetration of such devices within the scientific community. As mobile becomes mainstream for the less complicated functions commonly completed via a desktop computer, scientific software developers must actively work to provide equivalent functionality for mobile environments. Efforts by a small community of focused scientists already provide access to key data resources, chemical structure building and viewing applications, and molecular-modeling software, yielding expanded molecular-science toolkits for experts and novices alike.
Mobile database access
Because of the abundance of data on molecular entities, many investigations of chemicals start with simply ascertaining available knowledge. To that end, numerous apps provide focused sets of information (such as Green Solvents and TB Mobile). These relatively simplistic apps provide users with ways to explore collated sets of chemical information that are self-contained within the app.
Apps that act as a portal to large online databases allow for much broader use. Such apps provide a significant advantage over the corresponding website by solving the problem of allowing users to draw chemical structures for use as queries, as well as leveraging the improved performance and functionality of a native app for browsing the results. Apps, such as ChemSpider Mobile, SPRESImobile, and Mobile Reagents, are tied to a specific service. Apps, such as the Mobile Molecular DataSheet (MMDS) and MolPrime+, provide searching as a secondary feature and facilitate access to PubChem, ChEBI, and ChemSpider, in addition to their primary structure-data management capabilities. Such apps provide users with access to millions of chemical structures online with precomputed properties, associated experimental data, and links to additional data across the Internet. In the case of the ChemSpider Mobile app, more than 28 million chemicals can be accessed online with linked information, including patents, publications, chemical vendors, and a myriad of other related data types. The open availability of the ChemSpider web services allows for the pairing of diverse interfaces, including the mobile-app client, making use of the server-based searching and data storage (Figs. 1 and 2).
Apps are available for sketching and visualization of two-dimensional (2D) structures, three-dimensional (3D) structural models, molecular properties, and collective trends such as structure-activity analysis. Many of the apps currently available primarily are visualization tools, but increasingly they are able to link up with web services to generate models from user-provided data.
Mobile molecular collaboration platform
Mobile apps have access to strong collaboration features, owing to their origins as portable communications devices. Because apps are typically quite modular relative to desktop software, a workflow often operates by passing documents between different apps, and between heterogeneous platforms by sending and receiving e-mail attachments, downloading data from the Internet, and interacting with online storage services such as Dropbox.
It is also possible to connect with services, such as Twitter, for chemical data, images, and commentary. Using specific hash tags corresponding to topics of interest is a way to publish data to projects in an app, such as Open Drug Discovery Teams, which has an app-based interface for presenting scientific data on rare and neglected diseases as well as other research topics of interest; it is also chemistry aware, as shown in Fig. 3. Many of these apps can be used in an integrated workflow as data is passed from one app to another (see this article).
Mobile systems modeling
The interactions between chemicals and their environments have long been the domain of dynamics simulation software (such as MATLAB®). Because pharmacokinetic and pharmacodynamic models are typically crafted in the form of ordinary differential equations (ODEs), any ODE solver can theoretically be used to run such models. Coding an ODE solver is not complicated and several have been made available on both the Google Play store and iTunes App Store. As computational power in mobile devices increases, it may well be that developing more complex models that use these software tools is the next frontier for the mobile molecular sciences. Math Minion is a particularly well-developed example of the mobile capabilities for generating and running a mobile physiologically based pharmacokinetic (PBPK) model.
Math Minion provides users with a graphical interface for formulating models. The interface allows the definition of all relevant variables in a dynamic simulation and the necessary ODEs to be simulated. In addition, Math Minion will automatically analyze the units defined for all parameters, make conversions, and note inconsistencies. Math Minion has been used to implement a functional model of styrene pharmacokinetics (Fig. 4).
With such modeling function freely available, models for use in preclinical and clinical environments are sure to follow. Such models could direct scientists in pharmaceutical development on the proper dosing for animal studies, inform doctors on how to optimize radio-contrasting agents in radiology or anesthetics prior to surgery, and protect field risk assessors (for example, abandoned or active hazardous waste sites; environmental risk assessment; waste management) by allowing them to quickly determine if an environmental exposure is likely to cause high internal doses. A framework such as Math Minion can enable the development of dynamic system models that detail the influence of a chemical on its environment over time and feed into a plethora of other realms of the molecular sciences.
Server-side and cloud applications
In addition to the numerous client-side mobile applications that can be used in molecular sciences, many server-side web-enabled applications can be used to round out missing functionalities. Although these fall between mobile and traditional formats (often allowing access via both mediums), they are accessible via mobile devices using focused app interfaces or through web browsers. Key abilities that may not be otherwise accessible on mobile platforms because of computational or storage limitations can still be accessed on mobile devices as long as the necessary infrastructure is available elsewhere. Many tools (including some of those mentioned above such as ChemSpider Mobile) rely on the connected nature of mobile devices to provide users with the functionality they need on the devices they desire.
While many of these tools have interfaces specially formatted for mobile devices that provide a seamless integrated feel (for example, apps that link to data sources or mobile formatted websites), there are many tools that were originally designed for traditional computing environments that can outperform the current mobile tools. For example, to examine the interactions between a set of compounds and a protein of interest, several servers (for example, Dock Blaster, SwissDock, Docking Server) provide mobile modelers with access to protein-ligand docking capabilities. Web-based toolkits for application or generation of statistical quantitative structure-activity relationship (QSAR) models also are available (for example, Chembench and OCHEM). Because many of these web-based tools are designed for desktop environments and require formatted file inputs, file manipulation applications and/or nonnative web browsers with file-system access functions are required.
Although many tools exist for scientific mobile apps, few are tailored exclusively to the mobile modeler in comparison to the offerings in a desktop environment, leaving much room for further development. While it has been documented that rather complex molecular-science investigations can be completed by combining several apps, integrated research environments still remain elusive in the mobile environment. However, the recent inclusion of Windows 8 in new mobile devices blurs the distinctions between traditional computing and mobile computing. The expansion of molecular-science software available to scientists working exclusively on mobile devices will continue to accelerate.
[Disclaimer: The United States Environmental Protection Agency through its Office of Research and Development funded and managed the research described here. It has been subjected to Agency review and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.]