Mir Tanveerul Hassan

Baekdong 2-gil, Jeonju, Jellobuk-do, South Korea· (+82) 10-6523-3955. mirtanveer@jbnu.ac.kr

I am a passionate bioinformatics enthusiast focused on the analysis and prediction of therapeutic peptides using advanced machine learning (ML) and deep learning models. My work involves developing innovative computational methods to identify bioactive peptides with therapeutic potential. In addition to my research, I am highly involved in dataset preparation and preprocessing, ensuring data quality and consistency for robust model training and prediction accuracy. I am committed to leveraging cutting-edge technologies to advance the field of bioinformatics and contribute to the development of novel therapeutic strategies.

Education

Jeonbuk National University, South Korea

Ph.D.

Dept: Electronics and Information Engineering

Supervisior: Dr. Kil To Chong

Thesis: Machine Learning Platform for In Silico Analysis and Prediction of Therapeutic Peptides

GPA: 4.17/4.50

March 2021 - December 2024 (expected)

University of Kashmir, India

M.Tech

Dept: Computer Science

Supervisior: Dr. Rana Hashmy

Thesis: Facial Expression Detection Model using Keras

GPA: 8.85/10.00

October 2017 - September 2020

Islamic University of Science and Technology, Kashmir

B.Tech.

Dept: Computer Science and Engineering

Supervisior: Mr. Syed Mujtiba Hussain

Thesis: Detection of various athlete activities from video sources using MATLAB

GPA: 8.23/10.00

August 2012 - September 2016

Publications

Journal Publications

iAnOxPep: a machine learning model for the identification of anti-oxidative peptides using ensemble learning

Mir Tanveerul Hassan, Hilal Tayara, Kil To Chong

IEEE/ACM Transactions on Computational Biology and Bioinformatics (2024)
Possum: identification and interpretation of potassium ion inhibitors using probabilistic feature vectors

Mir Tanveerul Hassan, Hilal Tayara, Kil To Chong

Archives of Toxicology (2024): 1-11
NaII-Pred: An ensemble-learning framework for the identification and interpretation of sodium ion inhibitors by fusing multiple feature representation

Mir Tanveerul Hassan, Hilal Tayara, Kil To Chong

Computers in biology and medicine, 178 (2024): 108737
An integrative machine learning model for the identification of tumor T-cell antigens

Mir Tanveerul Hassan, Hilal Tayara, Kil To Chong

BioSystems 237 (2024): 105177
IF-AIP: a machine learning method for the identification of anti-inflammatory peptides using multi-feature fusion strategy

Mir Tanveerul Hassan, Saima Gaffar, Hilal Tayara, Kil To Chong

Computers in biology and medicine, 168 (2024): 107724
Meta-IL4: An ensemble learning approach for IL-4-inducing peptide prediction

Mir Tanveerul Hassan, Hilal Tayara, Kil To Chong

Methods 217 (2023): 49-56.

Conference Proceedings

Hybrid image fusion method based on discrete wavelet transform (DWT), principal component analysis (PCA) and guided filter

Andleeb Noor, Saima Gaffar, Mir Tanveerul Hassan, Mir Junaid, Aabid Mir, Amandeep Kaur

2020 First International Conference of Smart Systems and Emerging Technologies (SMARTTECH), IEEE, 2020
iTTCA: A sequence based predictor using feature representation for identifying Tumor T cell antigens

Mir Tanveerul Hassan, Kil To Chong

Journal of Information and Control (2022): 347-348
Aqueous Solubility Prediction Using Deep Learning

Mir Tanveerul Hassan, Kil To Chong

The Korean Institute of Electrical Engineers Conference (2022): 46-47

Projects

Anti-oxidative peptide identification & interpretation

Antioxidant peptides (AOPs) are promising natural alternatives to synthetic antioxidants. However, traditional methods for identifying AOPs are time-consuming. To address this, we have developed a novel computational predictor called iAnOxPep. This model utilizes ensemble learning techniques to predict AOPs based on their amino acid sequences. By overcoming limitations in existing methods, iAnOxPep offers a more efficient and accurate approach for identifying potential AOPs. This tool has the potential to accelerate the discovery of new AOPs and contribute to the development of innovative health products.

View on GitHub

Anti-inflammatory peptide prediction

Anti-inflammatory peptides (AIPs) offer a promising alternative to traditional anti-inflammatory drugs with fewer side effects. However, identifying these peptides remains a challenge. This study proposes a novel model, IF-AIP, to address this issue. The model utilizes a voting classifier and incorporates various feature descriptors to improve accuracy. When tested on independent datasets, IF-AIP outperformed existing methods in terms of accuracy and MCC scores. Furthermore, the model successfully identified all 24 novel peptide sequences as AIPs. These findings highlight the potential of IF-AIP as a valuable tool for the discovery of new AIPs and the development of effective treatments for inflammatory diseases.

View on GitHub

Sodium ion inhibitor prediction and interpretation

High-affinity ligand peptides that can control ion flow across cell membranes are crucial for treating various diseases. However, identifying these peptides remains a significant challenge. In this study, researchers developed a novel ensemble-based model called NaII-Pred to predict sodium ion inhibitors. This model utilizes six different descriptors and a hybrid feature set to train multiple machine learning classifiers. By combining the strengths of these classifiers, NaII-Pred achieves superior performance compared to existing methods. This powerful tool has the potential to accelerate the discovery of new sodium ion inhibitors and contribute to the development of innovative therapies for a wide range of diseases.

View on GitHub

Tumor T-cell antigen identification and prediction

The global rise in cancer cases highlights the urgent need for advanced treatments, with immunotherapy offering a promising approach by utilizing the immune system to combat cancer. Central to this strategy is identifying tumor T-cell antigens (TTCAs), essential for immunotherapy development. This study presents TTCA-IF, an innovative machine learning-based framework for TTCA identification. TTCA-IF integrates ten feature encoding types with five machine learning classifiers, producing 150 baseline models. These outputs are further refined through meta-models using an ensemble approach, culminating in the TTCA-IF predictive model. TTCA-IF outperforms existing predictors and baseline models in accuracy. In testing with nine novel peptides, TTCA-IF correctly identified eight as TTCAs, showcasing its exceptional precision. As a powerful screening tool, TTCA-IF holds significant potential in advancing cancer immunotherapy by facilitating the identification of promising TTCAs, paving the way for more effective cancer treatments.

View on GitHub

Potassium ion ligand prediction and intrepretation

Potassium ion flow through cell membranes is vital for processes like hormone secretion, epithelial function, electrochemical gradient maintenance, and electrical impulse formation. Potassium ion inhibitors hold promise in treating conditions such as cancer, muscle weakness, renal dysfunction, endocrine disorders, and cardiac arrhythmia. Identifying these inhibitors is crucial for regulating ion channel activity. In this study, we introduce POSSUM, a meta-model designed for identifying potassium ion inhibitors. Using two datasets, we trained and evaluated the model with seven feature descriptors and five classifiers, resulting in 35 baseline models. The optimal models were selected based on the mean Gini index score, and the POSSUM method was trained on probabilistic feature vectors. POSSUM demonstrated superior performance compared to baseline models and existing methods across both datasets. As a powerful tool for identifying and screening potassium ion inhibitors, POSSUM has the potential to advance treatments for various diseases by targeting potassium ion flow.

View on GitHub

Meta-IL4 inducing peptide prediction

Interleukin-4 (IL-4) is a crucial cytokine involved in immune responses. It plays a pivotal role in T-cell differentiation and activation. To facilitate the discovery of peptides that induce IL-4, researchers have developed a novel ensemble learning model called Meta-IL4. This model leverages various feature encodings and a combination of machine learning algorithms to accurately predict IL-4-inducing peptides. The model achieved impressive performance, with a high accuracy of 90.70% and a strong correlation coefficient of 0.793. This advanced tool has the potential to accelerate the development of therapeutic peptides that can modulate immune responses and treat immune-related diseases.

View on GitHub

Skills

Languages

C++

Java

Python

MATLAB

Technologies

Git

AWS

Linux

Bash

Libraries

Scikit-Learn

PyTorch

Keras

TensorFlow

NumPy

Pandas

Matplotlib

Seaborn

Interests

Apart from being machine learning enthusiatic in the realm of bioinofrmatics , I enjoy most of my time being outdoors. I love playing cricket (loved by over 2 billion people).

When forced indoors, I follow a number of anime movies and television shows, I am an aspiring chef, and I spend a large amount of my free time exploring the latest technology advancements in machine learning.

Awards & Achievements

BK21 Fellowship, Brain Korea Fellowship, 2024-2025
BK21 Scholarship, Brain Korea Scholarship, 2022-2025
GATE Fellowship, Ministry of Minority Affaris (MoMA), India, 2018-2020
MoMA Fellowship, Ministry of Minority Affairs (MoMA), India, 2013-2016

Resume

Download my CV (PDF)