Good day, Dr. Vasavi C S. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

Please start by introducing your professional journey in academia.

Is my Internet clear for you?

Yeah.

Thank you.

You're welcome. Please start by introducing your professional journey in academia.

Sure. So I'm Doctor Vasvi CS currently working as Assistant professor senior in Amrita Vishwa Vidyapeetham in the department of Artificial Intelligence. So talking about my educational qualifications, I have completed my PhD from Vellore Institute of Technology in the year 2021 and I finished my masters and I graduated in 2010 from.

Amrita Vishwavidyapitham, Coimbatore.

And I finished my masters in biomedical engineering and my bachelors. I graduated in 2008 from Satyabhama Institute of Science and Technology in the field of bioinformatics. And with respect to my professional career. So currently, as I told before, I am working as an assistant professor senior in Amrita Vishwavidyapetham in Bangalore campus. And before joining Abrita, I was working as a project scientist that I have data.

I worked on a project and and also I have quite a good number of working experience with respect to teaching. During my tenure of PhD at VIT in Vellore, I worked as an assistant professor junior and also I have worked as a teaching and research research associate. So before joining Vellore, I even worked for Neural Islam Institute in Nagar Koil in the department of Biomedical Engineering. And prior to that I worked as a.

At triple IT Hyderabad, so where?

Guest lecturer in bioinformatics at Bardasan Institute of Science and Technology. So Bardasan University at Chitrapalli and with respect to my research experience, so before like even now, I have written a funded project and currently I'm working on deciphering the cryptic pockets from a protein structure. And also before that we have during my journey at.

IHOP Data Supply T Hyderabad We have curated plus 20K data set that is very useful for machine learning studies. So it is like we wanted to provide a protein ligand interaction data set with the help of molecular dynamics simulation so that we will be able to understand the.

Complete flexibility of the protein structures rather than having a static structures. And so we released that data set and it was published in Nature Scientific data. And even during my journey at PhD, at VIT, I have published quite a lot of papers with respect to my research and also with collaborative research works. So that's all about myself. Thank you.

Thank you for sharing that, Dr. Vasvi.

Yes.

Let’s start with your expertise in bioinformatics, specifically focusing on your specialization in medical microbiology. Can you elaborate on your experience and contributions in this domain?

So with respect to medical microbiology, what I could tell is like.

As I don't have much of vet lab experience, so as my career was much focused on computational bioinformatics, so I have worked on with few viruses and some enzymes. So probably that can be always correlated with microbiology. So that's how I have done and I have done like I've also worked on some.

Uh, simulations, molecular dynamics simulations also, if there is no protein structure available, I have modeled the protein structures. Uh, so these are the things that I have worked and also like few docking studies when I collaborated with experimental people. So for whatever studies they are trying to have some experimental validations and insights if they have to understand. With respect to computational perspective and atomic level, what are the changes? And there I have contributed in terms of.

Medical microbiology. So that's all I could tell for now.

Bye bye.

Just to confirm, are you requesting to end the interview? Please clarify explicitly.

No, no, I'm not requesting for the end of interview. Yes.

Got it, thank you. Let’s move to the next area of expertise: your ability to teach both theory and laboratory courses. Could you share how you've designed or conducted courses that blend theoretical concepts with hands-on laboratory sessions?

Sure. So like with my experience at Amrita Vishwavidya Pitham. So the time I joined we were the first you know like batch and meaning like we were the first few faculties because the department was started the time I joined in 2023. So I was given a course like about interaction to biological data. So in that course I had. Like I had to teach.

Regarding like biological data for AI students, which which is quite challenging because I have a crowd where people are not interested with respect to biology. So I had a challenge where I need to make them understand the theory as well as I have to take it in a way where they are quite interested with respect to programming. So now what I did is like. I taught them a very basic theory.

Where they can understand quite a few concepts and I didn't get into so deep because whatever was required for programming, for example, I explained them what is protein structures and what are the databases that can have the protein structures. What do you mean by gene, gene sequences and which database could hold the gene sequence? And the interesting part is like I wanted them to decipher all this using some programming languages. So I.

To teach Python And I did shell scripting and also like TK console languages. So with the help of these programming knowledges and also with Linux I taught them like.

How to interpret a biological data and also how you can curate these kind of data set? Because the problem with respect to this particular biological field is that though we are talking in terms ahead of having AI in biology and all that, when you talk about any machine learning model or deep learning model, the first prior theme or the heart and soul of any biological thing is like you need to have a data set.

So the challenge comes there where we didn't have sufficient data set and any biologists like generally people try to take it from Kaggle and so on websites, but we didn't have anything like that. So the challenge comes where there where you had to curate your own data set.

So I taught my students how to curate your home data set and using Python programs how you can try to calculate the features.

In all that aspect with respect to protein structures or sequences or gene sequences, and then slowly with the help of those datasets, they try to implement machine learning models and that's how they did. So I think this is where I wanted to highlight where for a student who does, who are not passionate about biology. But they are more focused in understanding the programming.

I had to balance between both the crowds and somehow later I really got a good feedback where they were interested to work further for all these projects. So this is how I did it in Amrita and I would like to share that experience.

It’s impressive how you tailored your teaching approach to engage students with different interests, blending biology and programming effectively. Transitioning from theory to practical application—such as curating biological datasets and implementing machine learning models—highlights your ability to innovate and inspire. Let’s discuss your experience in student evaluation and examination duties. Could you share how you structured evaluation methods to ensure fairness and that students effectively grasped the material?

Yes, uh, yes, sure. So like in like for now, because since I'm working here, I would like to go with that pattern of evaluation with whatever we did. So we had 70% for internals and 30% for theory. I mean the for the end semester. So in the 70% internals we had. Quizzes, lab evaluations and also we can have like apart. These are like some common things that.

We generally do so apart from that, I really wanted the student to have a feel of how they are understanding the subject. So what I did is I made my end semester like I gave 30%, not 30%, like almost close to 45 percentage of weight age for their practical implementation. So where I was given, I gave them a privilege to. Choose a biological problem.

And I asked them to curate our data set. And so initially I used to teach them some programmings on how do you how how you can curate a data set like what is meant by translation, transcription and what are the things that you can look over it. So based on that, I try to make them understand and then I gave them.

Some idea over that and I asked them to have some practical implementation and because like I was teaching, but it's not like the theory, whatever we are teaching, the students tend to get you know like grasp the entire knowledge. So I wanted them to understand the practical perspectives of the subject as well. So in that case, apart from the regular lab classes.

So what I did is I made them to work for the project and further I evaluated them based on the practical knowledge along with the theoretical knowledge for the respective subject.

It’s clear that you’ve designed your evaluation methods to blend theory and practical application, giving students hands-on experience while ensuring they grasp the broader subject matter. Weighting practical projects heavily not only fosters deeper understanding but also equips students to tackle real-world challenges—a commendable approach. Next, let’s discuss your ability to guide student projects and research. Can you provide an example where you mentored a student or group, assisting them in navigating a complex research problem?

Yes, I feel whatever because.

Probably the problem.

Uh, like initial stage was not very complex, but when they started to work on then, uh, the complexity of problem comes. So, uh, uh yeah. So I could talk about one group where they wanted to implement some deep learning approach for understanding.

Uh, HIV protease, since it's my research fee. So they wanted to understand like, uh, what are the drug resistance mutations? So provided a sequence, uh, they wanted to know whether that sequence carries any drug resistance mutations, if it carries to which kind of drug it is resistant to. It's not only with respect specific to a specific, uh, protease enzyme. Uh, we had enzyme, different enzymes like reverse.

Script plays and all that. So they came up with such kind of problem, but then they weren't sure like from there to take the data set from or how to go about it, how to understand the resistance mechanism. And it's impossible for us to, you know, completely rely on, you know, like.

Literature reading it 1 by 1. So what I told them is like first initially you can have some NLP models where you can try to retry the.

Mutations and also check for some databases like Stanford databases gives good information on drug resistance profile. But the challenge there is you need to go and individually see whether that particular mutations is resistant to specific drug or not. So rather we wanted to give a sequence and say like for example if a person has.

Identified a sequence, so HIV sequence.

And they wanted to understand whether this particular sequence has any mutation or not. So I help them in understanding what are the mutations, what are the drugs and also I help them to.

Understand the problem statement in depth so that they can come up with good hybrid models where they can decipher the resistances for any given sequence. So this is one thing I could say.

And another thing is like with respect to my current PhD scholar, uh, she's from completely from, uh, computer science background where she doesn't have any knowledge on biology. So mentoring her also is like, uh, I won't, uh, it is like, uh.

I wouldn't say it is a challenge, but I feel it is a privilege where I can teach somebody where they're not aware and I can also help them to understand why it is important rather than going in a pure computer science field where you can also help in healthcare domain. So these are some examples where I try to help people and also with respect to my project which I had in the past 20 K data set. The one.

Which I did during my tenure that I have data. So there the challenge was building our data set. So it is not a simple data set we needed to build.

A trajectory like we wanted to understand a protein ligand interactions and it was not a static data, it is a dynamic data. So we subjected simulations for such kind of protein ligand in data set. So we had chosen.

Around 20,000 protein ligand complexes and subjecting each protein lichens complex for five runs.

So and then further we did binding free energy calculations to understand the stability of the protein like and complexes. So there the challenge was to curate 1,00,000 trajectories and make sure that everything is on place and further you know like like preparing the data set for machine learning studies so.

I had to handle a very big team. I we almost had some 25 interns because curating such kind of data set was a very great challenge. So there also I had a difficulty where a new students with like who doesn't have much of practical knowledge comes then I trained they didn't even have knowledge on Linux, so I had taken several sessions. So these 25.

Didn't join at a stretch so throughout the period they take turns like some may stay for six months then go and then some new students come so each time I had to teach them what is Linux, what is protein like interactions and how do you do molecular dynamics simulations and how do you.

Understand the binding free energies. So these are the challenges that I had and yeah, I successfully completed and all these things, whatever I'm saying, except for the PhD work, it is all reflected in the terms of publications as well.

You’ve clearly demonstrated strong mentorship skills, guiding students and research scholars through complex projects while addressing their varied levels of expertise. From explaining HIV drug resistance mutations and developing hybrid models, to leading a large-scale data curation team for protein-ligand interactions, your ability to manage challenging projects and diverse groups is evident. Let’s move to communication and teaching. How do you ensure your approach is structured and clear enough to cater to both beginners and advanced learners in your classroom?

OK.

Sorry to interrupt, we're down to the last 5 minutes.

So.

For uh, so whenever I start teaching uh, uh, like I have uh.

Like the crowd, uh, all, most of the time had been at a beginner level, so I had to teach them with the basics and then slowly.

Move ahead to the advanced approaches.

I think.

The examples what I had told earlier. So I had a crowd where I had trained a set of candidates and also some candidates were new who wanted to learn where I was not reluctant. I was always motivated though. It was like something that has to be repeated again and again for a new set of crowd I always had. Happiness and pleasure. OK, I'm teaching something new.

For those people.

Who are not?

Much sorry.

Who who, who doesn't have much of idea so.

For the beginners, I make sure, uh, that they understand and, uh, even I go and talk to them personally as well during class hours. And also I asked them to come and meet me, uh, post class. I was, if they have any doubts. And if I find that the student is not, uh, you know, like, uh, it's not, he's not able to understand. I ask him to come at his convenient time and make sure that he understands. That's all I have quite a lot of.

Experience on this lake during my tenure here and the students also felt happy and.

Like the challenges, like people who never liked the subject also started liking it, who also found it hard. It's not that every time I was able to manage 100%. Obviously depending on the student's level, there were students who were.

Who were not able to.

Grasp it completely and how much ever I try to help them, they were not ready to accept. I tried to. Apart from telling this I always I also motivate the students. Don't think that that the subject will not work for you. Always think that you will be able to do it and take it in a positive attitude. Don't be don't feel negative about yourself. So apart from teaching, I also help the students to get self motivated.

So that they will be able to progress ahead.

Your effort to address students’ individual needs, including spending time with them personally and motivating them, highlights your commitment to education and teaching excellence. Balancing foundational concepts for beginners while gradually introducing advanced topics to keep everyone engaged reflects a structured and student-centered teaching philosophy. Now, shifting focus to your PhD and publications—can you share how your doctoral research and published work contribute to the field of bioinformatics?

Sure, so my my PhD work was to understand the drug or specify the drug resistance mechanism in circulatory recommend form A1 HIV protease. So as we know, HIV is one of the according to WHO, it is one of the top diseases in the world where people. Though there are a lot of medications still, we are not able to find a complete cure and a lot of mortality rates are also increasing.

So what we wanted to understand is that how we can.

In spite of FTA approved drugs, we still found that these drugs were not effective against these proteins. So we took a specific target called as HIV 1 Proteas and what we did is we wanted to understand the drug resistance mechanism at the atomic level while drugs are not being effective. So then slowly when we started then we understood that there are different types of subtypes and it is not only the mutations that are causing the resistance and.

These subtypes differ individually for each country. Like for example subtype C is more prevalent in India and certain subtypes are more prevalent in Asia. So now they have some naturally occurring polymorphisms allowed in them. And now when drug resistance mutations along with this naturally polymorphisms when they Co occur, what happens is they drastically affect the binding of the inhibitor. So with respect to different subtypes it's been observed that.

The existing protease inhibitors have very weaker binding affinity to this existing protease enzymes. Uh, so the reason is that because of this naturally occurring polymorphism and also the existing protease inhibitors were mainly designed on based on subtype B protease. So now when we talk about other subtypes, as I mentioned that they have naturally occurring polymorphism. So obviously because of this polymers polymorphism, uh, the protease inhibitors tended to have a weaker interactions, which we could understand from the literature.

And now we wanted to see like like there were quite a few papers also reporting that when mutations were Co occurring, it caused caused resistance. So we wanted to know the mechanism in which this resistance is being caused. And so we did molecular dynamics simulations and binding free energy calculations to understand such kind of resistances. And we hope that whatever information we decipher could be helpful.

In designing new inhibitors.

Uh, or tailored design inhibitors for specific subtypes, uh, more, uh, not only generally focusing on enzymes, but, uh, we wanted to highlight apart from mutations and all this subtype, taking into consideration of subtypes is also equally important.

Your PhD research offers significant insights into the nuanced drug resistance mechanisms in HIV protease, particularly highlighting the interplay of naturally occurring polymorphisms and drug resistance mutations across subtypes. This could undeniably aid in the development of tailored inhibitors for diverse subtypes, addressing a crucial gap in HIV treatment strategies globally. Are there any final topics or aspects of your background you’d like to highlight before we wrap up?

Uh, no. I think I have uh, told enough about myself so.

That's all.

Thank you for sharing your expertise and experiences, Dr. Vasavi. To confirm, would you like to officially conclude the interview now?

Yes.