Donald P. Visco

Developing a methodology for an inverse quantitative structure-activity relationship using the signature molecular descriptor.

The concept of signature as a molecular descriptor is introduced and various topological indices used in quantitative structure-activity relationships (QSARs) are expressed as functions of the new descriptor. The effectiveness of signature versus commonly used descriptors in QSAR analysis is demonstrated by correlating the activities of 121 HIV-1 protease inhibitors. Our approach to the inverse-QSAR problem consists of first finding the optimum sets of descriptor values best matching a target activity and then generating a focused library of candidate structures from the solution set of descriptor values. Both steps are facilitated by the use of signature.

Computer-aided molecular design using the Signature molecular descriptor: Application to solvent selection

Abstract There is a growing demand to develop more environmentally friendly solvents to reduce costs and comply with regulation. Researchers at GlaxoSmithKline (GSK) have developed a solvent selection guide that ranks 47 frequently used solvents from 1 to 10 in five areas related to environmental compatibility. In this work, we apply a computer-aided molecular design method known as inverse design with the Signature molecular descriptor to identify additional potentially green solvents outside of GSKs list. Applying this approach is much quicker, less expensive and allows for a more comprehensive search for the most suitable candidates than working with experimental data alone. We present results for solvents with optimal predicted properties that span the classes from the 47 compounds in the GSK solvent selection guide and include several which are hybrids that cross-cut amongst classes. Additionally, our technique “rediscovers” the known green solvent ethyl lactate through this method by combining different solvent classes.

Data mining PubChem using a support vector machine with the Signature molecular descriptor: classification of factor XIa inhibitors.

The amount of high-throughput screening (HTS) data readily available has significantly increased because of the PubChem project (http://pubchem.ncbi.nlm.nih.gov/). There is considerable opportunity for data mining of small molecules for a variety of biological systems using cheminformatic tools and the resources available through PubChem. In this work, we trained a support vector machine (SVM) classifier using the Signature molecular descriptor on factor XIa inhibitor HTS data. The optimal number of Signatures was selected by implementing a feature selection algorithm of highly correlated clusters. Our method included an improvement that allowed clusters to work together for accuracy improvement, where previous methods have scored clusters on an individual basis. The resulting model had a 10-fold cross-validation accuracy of 89%, and additional validation was provided by two independent test sets. We applied the SVM to rapidly predict activity for approximately 12 million compounds also deposited in PubChem. Confidence in these predictions was assessed by considering the number of Signatures within the training set range for a given compound, defined as the overlap metric. To further evaluate compounds identified as active by the SVM, docking studies were performed using AutoDock. A focused database of compounds predicted to be active was obtained with several of the compounds appreciably dissimilar to those used in training the SVM. This focused database is suitable for further study. The data mining technique presented here is not specific to factor XIa inhibitors, and could be applied to other bioassays in PubChem where one is looking to expand the search for small molecules as chemical probes.

Potential Glucocorticoid Receptor Ligands with Pulmonary Selectivity Using I‐QSAR with the Signature Molecular Descriptor

We intend in this research to establish a rational method for the development of novel glucocorticoid receptor ligands to more effectively prevent respiratory inflammation. Corticosteroids, a class of steroid hormones, are naturally inclined to bind to the glucocorticoid receptor and, in this research, are the basis for exploring other novel and non‐intuitive structures. To be more effective than currently available medications, novel compounds must be highly selective toward the lungs and must be inactivated when exposed to the main circulation, thus preventing the participation of the ligand in other systems and consequently reducing systemic side‐effects. We look to use the inverse‐quantitative structure–activity relationship algorithm with the Signature molecular descriptor to generate new ligands based upon the structures and activities of 65 experimentally studied corticosteroids. Inverse‐quantitative structure–activity relationship explore many possible combinations of atom connectivity while structural filters and other scoring approaches are used to predict and identify the most promising candidates for further study. Properties explored include high receptor binding affinity, high systemic clearance, high plasma protein binding and low oral bioavailability. Among more than 300 million potential candidates generated, 84 high priority compounds with properties predicted to be at least as or more effective than currently available corticosteroids have been identified with this procedure.

Evaluating the SAFT-VR and the Sanchez–Lacombe EOS for Modeling the Solubility of Blowing Agents in Polyols

The statistical associating fluid theory with variable range (SAFT-VR) is a robust equation of state (EOS) with six parameters that describe the thermodynamics of complex systems. Earlier works with this EOS have already predicted the phase co-existence properties of various refrigerants and higher order alkane series compounds, along with their mixtures. In this work, the SAFT-VR EOS has been used to correlate the P–V–T properties of three polyols, namely Pluracol 355, Pluracol 975, and Terol 352. The solubility of nine, zero ozone-depleting blowing agents in these three polyols has been correlated wherever the experimental data are available, and the solubility curves have been produced. The solubility results obtain from SAFT-VR EOS have been compared to a lattice theory based EOS known as Sanchez–Lacombe (SL). In addition, the solubility results of some blowing agents are predicted using the SAFT-VR EOS and the SL EOS, in the above-mentioned polyols wherever there is no experimental data. In general, the SL EOS predicts the solubility of the blowing agents in the polyols more accurately than the SAFT-VR EOS.

A proposed teaching and learning curriculum for COMPLEETE based on current national trends

We propose an introductory level teaching and learning curriculum for the ASEE COMPLEETE program (COMPetencies in Learning for Engineering and Engineering Technology Educators). COMPLEETE is an initiative for a national program to build and recognize educator excellence in engineering and engineering technology at three levels. The proposed curriculum for the introductory level is compared with curricula from nine well-established existing programs. The content is specifically targeted to benefit engineering and engineering technology instructors in higher education, integrate with the values and programs already offered within ASEE, serve as a foundation for further development at higher levels, and be flexible to suit the needs of a diverse instructional community. The nine existing programs were coded under the overarching COMPLEETE criteria and then analyzed for commonalities and alignment. The proposed core competency areas were found to comprehensively represent existing programs. They are: learning theory, student development, instructional design, instructional facilitation methods, assessing and providing feedback to learners, instructional technology, and reflective practice. The proposed curriculum lays a foundation for those offering faculty development services to compare against, and challenges the engineering and engineering technology community of educators to address key competency areas all faculty should develop within 3-5 years of beginning teaching.

Impulse Backscattering based Detection and Imaging of Shallow Buried Objects

We discuss our recent work on simulational studies of impulse propagation and backscattering for the detection and imaging of shallow buried objects in close packed granular beds.

Gas Diffusivity Through EPS Foams

A simple multiscale model was developed and used to predict gas diffusivities through expanded polystyrene foam at near standard temperature and pressure conditions. The technique involves measuring gas diffusivities at various length scales then combining them using an electrical analogy for parallel resistances to construct an effective property. A commonly used experimental technique, the continuous flow method, was used to obtain diffusivity data for argon through polystyrene films and foams. Although a simple Fickian mathematical model was able to predict diffusivities through films, a simple ‘coarse’ multiscale model that accounts for the morphological features was developed for the foam.

Acoustic interrogation of soil and possible remote detection of shallow buried inclusions

Here we address the problem of remotely interrogating the shallow subsurface of soil using low power mechanical energy transmission and sensing the subsequent backscattering from the soil bed to find small buried objects at very shallow depths (~ 15 cm or less). The effort is geared towards the development of better technologies for the remote detection and imaging of buried land mines, improvised explosive devices (IEDs) and other undesirable objects. We present our studies against the backdrop of what is known in this field today. Our ongoing work is briefly summarized with an outline of the underlying physics of acoustic backscattering from shallow buried objects and of how such backscattering can be inexpensively and remotely detected and interpreted for possible use in automated mine clearance operations.

Pharmaceutical Machine Learning: Virtual High-Throughput Screens Identifying Promising and Economical Small Molecule Inhibitors of Complement Factor C1s

When excessively activated, C1 is insufficiently regulated, which results in tissue damage. Such tissue damage causes the complement system to become further activated to remove the resulting tissue damage, and a vicious cycle of activation/tissue damage occurs. Current Food and Drug Administration approved treatments include supplemental recombinant C1 inhibitor, but these are extremely costly and a more economical solution is desired. In our work, we have utilized an existing data set of 136 compounds that have been previously tested for activity against C1. Using these compounds and the activity data, we have created models using principal component analysis, genetic algorithm, and support vector machine approaches to characterize activity. The models were then utilized to virtually screen the 72 million compound PubChem repository. This first round of virtual high-throughput screening identified many economical and promising inhibitor candidates, a subset of which was tested to validate their biological activity. These results were used to retrain the models and rescreen PubChem in a second round vHTS. Hit rates for the first round vHTS were 57%, while hit rates for the second round vHTS were 50%. Additional structure–property analysis was performed on the active and inactive compounds to identify interesting scaffolds for further investigation.