Meet Sahil: A Norway-based Computational Chemist Developing Molecular AI Skills at Neovarsity

In the picturesque city of Tromsø, Norway, where the midnight sun illuminates the landscape, we find Sahil Gahlawat, a talented researcher from India.

Sahil is on the cusp of completing his PhD in computational chemistry at UiT The Arctic University of Norway.

His research focuses on modeling homogeneous metal-catalyzed reactions, resulting in the co-authorship of several impactful papers.

Building on his expertise in computational chemistry, Sahil is expanding his skill set to include molecular machine learning through Neovarsity's advanced curricula in cheminformatics and machine learning for small molecules.

This additional training is proving to be a crucial component in rounding out his expertise, equipping him with the tools to apply modern data-driven approaches to small molecule research.

Sahil’s Research Journey from India to Norway

Sahil's journey began at the prestigious Indian Institute of Technology Roorkee, where he completed his Master of Science degree, focusing on computational screening of silicon-containing electrolytes for lithium-ion batteries.

This laid the foundation for his PhD pursuit in Norway.

During his doctoral studies, Sahil enhanced his expertise through two significant internships: at AstraZeneca in Gothenburg, gaining valuable industry experience, and at the University of Oslo, where he worked on diverse projects from ab initio molecular dynamics to predicting hydrogen-isotope exchange reactions.

We sat down with Sahil to discuss his research career, how Neovarsity's curricula are shaping his high-level skills and his career aspirations.

What motivated you to pursue a PhD in computational chemistry at UiT The Arctic University of Norway?

I worked as a synthetic chemist until my bachelor's. I was introduced to computational chemistry in my master’s. In my thesis work, I applied quantum chemical tools like DFT to predict the stability of silicon-based electrolytes for lithium-ion batteries. I was amazed by the power and potential of computational chemistry and decided to pursue a PhD in the same field. My love for nature, especially mountains, motivated me to apply for a PhD in Norway.

Can you elaborate on your current research work and its potential impact in the field?

My research involves predicting the reaction mechanisms of transition metal-catalyzed processes, with a focus on CO2 incorporation. I collaborated with experimental research groups at UiT, Yale, and Aarhus universities to publish collaborative work in impactful journals. I did a research exchange at AstraZeneca, where I trained a machine learning model for predicting the output of hydrogen-isotope labeling reactions, an important process for understanding the pharmacokinetics of a drug candidate. I performed data curation on the AstraZeneca database to extract training data and applied cheminformatics modules like RDKit to generate features for the machine learning model. Additionally, my research exchange at the University of Oslo allowed me to apply ab initio molecular dynamics simulations to enhance the accuracy of 19F NMR chemical shift values. I believe that my research work showcases the potential of computational chemistry in the field of catalysis and drug discovery.

What core computational chemistry skills have you developed during your PhD?

During my PhD, I developed skills for employing various quantum chemistry packages to perform high-quality static and molecular dynamics simulations for predicting molecular energies and other properties. I gained proficiency in the Python programming language as I constantly used it to generate scripts and implement the workflow of quantum chemistry packages on HPCs. I also acquired skills in cheminformatics and molecular machine learning as part of my research project with AstraZeneca.

You recently completed a course in Cheminformatics at Neovarsity. What motivated you to make this decision?

During my research exchange at AstraZeneca, I trained a machine learning model for predicting the output of hydrogen-isotope labeling reactions, an important process for understanding the pharmacokinetics of a drug candidate. I performed data curation on the AstraZeneca database to extract training data and applied cheminformatics modules like RDKit to generate features for the machine learning model. This project motivated me to enhance my skills in Cheminformatics, and I chose Neovaristy as I liked the advanced curriculum of the course, involving both theoretical and practical knowledge.

How has the Cheminformatics curriculum at Neovarsity complemented your existing skills?

During my training with Neovarsity, I have mastered cheminformatics as a powerful tool for designing small molecules, calculating varying molecular descriptors, including fingerprints, based on my system's properties, and integrating these into machine learning algorithms. I completed a project in drug discovery as part of the course, applying techniques such as Quantitative Structure-Activity Relationship (QSAR), clustering, combinatorial library design, and advanced machine learning algorithms.

You've further pursued advanced machine learning for drug discovery. What inspired this additional specialization?

The Cheminformatics curriculum allowed me to extract, analyze, and standardize molecular datasets, as well as generate varying molecular descriptors. The next step in the process will be to use the data to predict the desired output for the drug discovery process using machine learning algorithms. This inspired me to pursue advanced machine learning for drug discovery.

How is your journey with advanced machine learning progressing, and how do you see it enhancing your research capabilities?

My journey with advanced machine learning for drug discovery has been incredibly rewarding. The integration of machine learning techniques has significantly enhanced my research capabilities by allowing me to analyze vast datasets more efficiently and accurately. I am now able to identify potential drug candidates, predict their activity, and optimize lead compounds with greater precision. This has streamlined the drug discovery process, making it faster and more cost-effective. As I continue to explore new algorithms and methodologies, I am confident that these advancements will lead to groundbreaking discoveries and innovations in my research.

In what ways do you think the combination of your PhD work and Neovarsity training sets you apart in the field of computational chemistry and molecular machine learning?

The combination of my PhD work and Neovarsity training uniquely positions me at the forefront of computational chemistry and molecular machine learning. My PhD research has given me a deep understanding of homogeneous metal-catalyzed reactions, DFT calculations, and quantum chemistry, providing a solid foundation in theoretical and practical aspects of the field. Complementing this, my Neovarsity training has equipped me with advanced skills in cheminformatics, molecular descriptor calculation, and the application of machine learning algorithms. This dual expertise allows me to:

• Integrate traditional and modern techniques: Leverage classical computational chemistry methods alongside cutting-edge machine learning tools to solve complex problems in molecular design and drug discovery.
• Innovate in molecular modeling: Utilize cheminformatics and machine learning to develop predictive models, enhancing the accuracy and efficiency of molecular simulations.
• Optimize drug discovery: Apply QSAR, clustering, and combinatorial library design to streamline the drug discovery process, identifying promising candidates with higher precision.
• Contribute to interdisciplinary research: Bridge the gap between computational chemistry and artificial intelligence, fostering interdisciplinary collaborations and driving innovation.

How do you envision applying your unique skill set in your future career?

I envision applying my unique skill set in computational chemistry and molecular machine learning to innovate drug discovery, develop personalized medicine, and create sustainable chemical processes. By integrating advanced machine learning algorithms with my expertise in molecular descriptors and QSAR modeling, I aim to streamline the drug discovery process, optimize treatments based on individual profiles, and contribute to green chemistry solutions. Whether in academic or industrial research, I plan to lead cutting-edge projects and collaborate across disciplines, driving innovation and solving complex scientific problems. Additionally, I aspire to teach and mentor the next generation of scientists, sharing my knowledge to foster future leaders in the field.

Can you share an example of how you've already integrated or are planning to integrate machine learning techniques into your research?

I integrated machine learning in my research during my exchange at AstraZeneca, where I trained a machine learning model for predicting the output of hydrogen-isotope labeling reactions, an important process for understanding the pharmacokinetics of a drug candidate. I performed data curation on the AstraZeneca database to extract training data and applied cheminformatics modules like RDKit to generate features for the machine learning model. This project excites me, and I look forward to further opportunities for integrating machine learning techniques in my research.

What advice would you give to other computational chemists considering expanding their skills into machine learning?

Start by mastering the basics of machine learning, and then focus on integrating these techniques with your chemistry knowledge. Practical experience through projects and collaborations is key. Stay curious and keep up with the latest advancements in both fields. I recommend the curriculum from Neovarsity for expanding the skill set.

Note for Recruiters

Recruiters looking for a dynamic professional with Sahil's unique background in computational chemistry and growing expertise in molecular machine learning, please reach out. Sahil welcomes inquiries through us at [email protected] or via his LinkedIn profile.

You can explore Sahil's profiles to see more of his work:

LinkedIn: https://in.linkedin.com/in/sahil-gahlawat-509450bb

ResearchGate Profile: https://www.researchgate.net/profile/Sahil-Gahlawat

Google Scholar Profile: https://scholar.google.com/citations?user=FIYc9DAAAAAJ&hl=en