Final Report: Combining docking and free energy-based methods to improve virtual screening of drug candidates against HIV integrase

The title of this project is “Combining docking and free energy-based methods to improve virtual screening of drug candidates against HIV integrase.” Because HIV integrase is a key protein in the HIV-AIDS virus, the purpose of our research was to utilize docking, molecular dynamics simulation, and free energy calculation to computationally predict the success of HIV drug candidates in countering this protein. Similar to drug testing for other ailments, HIV drug testing and clinical trials typically require copious amounts of money, effort, and many, many years. With our work, we aim to use in-silico screening of drug candidates in order to focus the biomedical community’s time and resources on solely the most promising drug candidates for increased efficiency. Starting with software like Desmond from D.E. Shaw Research and then advancing towards using more powerful and advanced molecular modeling software, our research team collaborated in mastering techniques and determining the most promising HIV drug candidates. As our work continues to progress, we shall send our finalized results and data to Dr. Deng’s colleague at the University of Colorado for synthesis and further testing in a lab setting. Data analyses and results shall be published in an official scientific paper and also presented at the 2019 American Chemical Society national conference.

In-silico drug screening largely rested in the use and mastery of molecular docking. Molecular docking is the process of predicting ligand conformation and orientation within a targeted binding site. Docking aims to find the ideal interaction energy between a specific receptor and ligand in order to create an ideal and stable compound. Within the docking software, we can compare interaction energies of different poses by using the software’s scoring function. For each possible pose, considering all conformational degrees of freedom, the scoring function gives a value that helps us evaluate its affinity. To accurately simulate these interactions as they occur within the human body, we instructed the software to create a body of water molecules surrounding the protein of interest. Using Command Line Interface (CLI) within Linux, we simulated how the protein exists in solution and proceeded to make observations and collect data regarding hydrogen bond strengths and other interactions between atoms. In particular, we shall continue by analyzing binding energy values and considering enthalpy, entropy, and other factors. Our team has not yet produced sufficient results for proper presentation, nonetheless we, under the tutelage of Dr. Deng, already have steady plans for the year ahead to advance our research using more advanced methods within the software as well as free-energy calculations in order to collect, finalize, and publish results in a timely manner.

The level of detail and intricacy that docking and molecular modeling require have made our work rather challenging. In the earliest stages, there were already issues with the software Desmond as it failed several simulations we tried to run. After troubleshooting and investigating possible sources of error, it became clear that there were some problems within the software itself that caused some simulations to “die.” Fortunately, Dr. Deng had planned on using Desmond for only the first few steps of our learning process, so we promptly switched to more advanced software. However, we must continue to work with caution and precision as similar issues and challenges will always arise within this type of research. Overall, mastering the various functions, formatting, and even the CLI within Linux proved challenging, and I am determined to continue putting in time and work to improve and excel.

My time conducting research with Dr. Deng this summer has enriched me through the lessons I learned and the challenges I encountered. This was the first time I ever delved into real, in-depth research, and I have been incredibly grateful for the experience. As a biochemistry major who excels in mathematics and biochemistry but has rather limited preliminary knowledge of coding and computing, I initially found this type of research somewhat intimidating. However, Dr. Deng has been excellent as a mentor and guide in teaching me about conducting good research, CLI, and the vast array of possibilities that molecular modeling presents as it relates to medicine and chemistry. I now find this field of academics to be exciting and intriguing, and the skills I have gained in problem-solving, coding, and modeling will surely be very helpful both now and in the future. Although I am currently a student on the pre-med track with plans of becoming a physician, I am now open to the idea of research not only in a lab setting but also from a more computationally-focused context. Pace University’s undergraduate student-faculty research program has given me a remarkable opportunity to enrich my knowledge and reinforce my desires to contribute to the world as part of the medical community; in particular, I now plan to continue conducting this research with Dr. Deng and likely take on similar medicine-oriented projects later in my academic career.

I am very eager to continue working with Dr. Deng and our team to collaborate, analyze, and finally see our work come to fruition in the future. It is a unique and splendid opportunity for a (formerly) first-year student like me to be given the chance to contribute my talents to important research so soon as the summer after freshman year. I have always dreamed of serving the community and changing lives as a physician, and I am extremely grateful to be able to get involved with medicine, contribute my talents towards the process of finding a cure for HIV-AIDS, and overall find a small way for me to potentially give back to the world. As we continue this project, I am determined to continue learning and to contribute to this incredibly important research in hopes of finding a cure for the over 30 million HIV-AIDS victims worldwide.

Updates: Combining Docking and Free Energy-Based Methods to Improve Virtual Screening of Drug Candidates Against HIV Integrase

Research done by Dr. Nanjie Deng, myself, and our team this summer has made steady progress. As described in my previous blog post, our research revolves around the use of docking, molecular dynamics simulation, and free energy calculation to improve computational screening of potential HIV-AIDS drug candidates. So far, Dr. Deng has placed emphasis on preliminary research and the usage of Desmond, a software package from D.E. Shaw Research designed to perform molecular dynamics for biological molecules such as those of interest within our research.

In order to accurately predict interactions between drug candidate compounds and HIV integrase (a key protein in the HIV-AIDS virus) within the human body, we first must simulate how a protein exists in solution. Chemical reactions within the human body exist in solution, thus we used Desmond to simulate a “box” of water molecules around the protein we wanted to observe. Doing so using Command Line Interface (CLI) on a computer with Linux, we observed the strengths of hydrogen bonds and other interactions between various atoms (oxygen, nitrogen, carbon, hydrogen, etc.). As research advances, we will be using other 3D shapes of water to surround the protein; this will minimize the number of water molecules required within calculations and, consequently, increase efficiency. According to Dr. Deng, we will be reporting official data soon as our research continues throughout the academic school year and even into next summer. Our data reports shall be sent to a close colleague of Dr. Deng for further testing in a lab setting.

There have been many challenges during our research. Desmond proved to have some issues within its software, seeing that certain simulations failed to proceed, or “died,” and no mistakes on our part were found. This may be due to the fact that Desmond is a free software, and this is why Dr. Deng had already planned to only use Desmond for these beginning stages. Hence, more demanding and specific research shall be executed with more advanced, professional, and reliable molecular dynamics software. However, similar issues can exist in any software, so we must learn to use problem-solving skills to fix any future problems we may encounter. Another challenge I, personally, have encountered is that of learning CLI within Linux. I am a biochemistry major who has had very limited experience in coding. I will need to spend a significant amount of time learning this, but I am grateful; this is a very valuable skill that is utilized in many fields of study. Despite these challenges, I am happy to say that our team has made good progress under the tutelage of Dr. Deng.

Research so far has greatly broadened my horizons and reinforced my dreams of studying medicine and serving the community as a physician. I have always had success and been interested in biochemistry and medicine, but now I have been given the opportunity to learn about the application of coding and computing within the medical field. Not only will learning coding be an enriching experience and a vital addition to my education, but it will also prepare me for medical school and my future career. Working together as a team to achieve new goals has given me insight on how future work in the medical field will be. This work is incredibly important as it serves to help modern medicine find a cure for one of the world’s most merciless diseases, and I am motivated to contribute my knowledge and efforts towards making a change. I did not expect to have such a great opportunity this early in my life to get involved in medicine and the pursuit of new treatments/cures, and I am incredibly appreciative. I shall put in whatever time and work it takes to help the world become at least one step closer to curing HIV-AIDS.

Combining Docking and Free Energy-Based Methods to Improve Virtual Screening of Drug Candidates Against HIV Integrase

HIV-AIDS is a devastating disease that has a great impact on over 30 million people worldwide. A key protein in the virus is HIV integrase, and promising drugs to counter this disease should target this particular protein. Thus, under the guidance and leadership of Professor Nanjie Deng, the purpose of our team’s research is to collaboratively improve computational screening of drug candidates countering HIV integrase. Together with Dr. Deng, we shall utilize docking, molecular dynamics simulation, and free energy calculation to determine which drug molecules have the highest chances of success. Once we determine the most promising compounds, our results shall be sent to the University of Colorado so that Dr. Deng’s collaborator Professor Kvaratskhelia may synthesize these compounds in his lab for further testing. These compounds and published results shall be presented at the 2019 American Chemical Society national meeting.

Our research particularly aims to generate predictions of molecular compounds that may best bind HIV integrase. Consequently, the different methods and processes required for our research are key. Molecular docking is the process of predicting ligand conformation and orientation within a targeted binding site. Docking aims to find the ideal interaction energy between a specific receptor and ligand in order to create an ideal, stable resulting compound. Within the docking software, one can compare interaction energies of different forms/poses by using the software’s scoring function. For each possible pose, considering all conformational degrees of freedom, the scoring function gives a value that helps the researcher evaluate its affinity. We can increase efficiency by keeping the receptor rigid while keeping the ligand flexible; as a result, possible poses are more limited, and researchers have a narrowed down array of positions to evalulate. In general, this leads to researchers considering the binding energy values among other factors (e.g., enthalpy and entropy). Docking requires a rigorous amount of detail and specificity, however, thus this process can prove to be challenging. My particular role in our research lies in learning to run molecular dynamics simulation calculations in order to better observe and analyze binding for data collection.

We are determined to use our research to increase the efficiency and  success of HIV drug testing and production. I expect that the predicted compounds in our findings shall serve as potential potent HIV drug candidates, and we shall work to create results that could ultimately lead to the betterment of HIV treatment for victims of the deadly ailment around the world. With our research’s predictions and data, the clinical trial process for HIV drug candidates can be made most efficient since funds and time shall be focused on solely testing the very few, most promising candidates. Throughout my time in this research project, I hope to contribute my talents to our team’s success and immerse myself in a medical research environment; specifically, this shall enrich my knowledge and prepare myself for medical school and a future in the medical field and research. Furthermore, our research shall demonstrate the potential and utility of docking, molecular dynamics simulation, and free energy calculation for in-silico screening of drug candidates. For the benefit of mankind, we shall use our work to reveal modern medicine to a new set of possibilities and potential in drug production.