Md Abid Hasan

Ph.D. Advisor: Prof. Stefano Lonardi

Algorithms and Computational Biology Lab
Department of Computer Science and Engineering
Engineering Building Unit II, Room 362
University of California
Riverside CA 92521



About me

I am Md Abid Hasan. I prefer Abid. I was raised in Dhaka, Bangladesh. I obtained my Bachelors and Masters in Computer Science and Enginnering from Islamic University of Technology (IUT). It is one of the most prestigious engineering universities in Bangladesh. After graduating top of my class, I worked as a lecturer at the same university. During this period I completed my masters. In 2014, I moved to USA for my Ph.D. at University of California Riverside. Currently I am a fifth-year Ph.D. student at the Algorithms and Computational Biology lab. On summer 2018, I had the opportunity to work as a Bioinformatics intern at Roche Sequencing Solutions at Pleasanton, CA. My research interests are Computational Biology, Machine learning and statistical model in bioinformatics, Cancer genomics, and Deep learning in bioinformatics.

I will be working as a Principal Scientist II at Roche Sequencing Solutions, Pleasanton, CA starting December, 2019.

Current Research

Epi2En: A Convolutional Neural Network for Genome-wide Enhancer Prediction

Submitted to BMC Genomics

Enhancers are distal cis-regulatory elements that play a critical role in gene expression control. Their remote location relative to the target gene(s), their complex functional properties, and their lack of discriminating motifs and evolutionary-conserved sequence makes it challenging to detect them. The availability of genome-wide epigenetics data for a variety of cell lines has opened the possibility for a machine learning approach for the identification of enhancer regions. We propose a convolutional neural network called Epi2En for the prediction of enhancers across multiple human cell-lines from twelve epigenetic features.

Fig: Epi2En model architecture

DeeplyEssential: A Deep Neural Network for Predicting Essential Genes in Microbes

Accepted for publication at BMC Bioinformatics

Essential genes are critical for the survival of an organism. The prediction of essential genes in bacteria can provide targets for the design of novel antibiotic compounds or antimicrobial strategies. We have built a deep neural network classifier (DeeplyEssential) to classify essential and non-essential genes in these bacteria species. Our model only uses sequence-based features to predict the gene essentiality. Moreover we also expose and study a hidden performance bias that affected the previous classifier.

Fig: DeeplyEssential model for gene essentiality prediction

HiC interaction prediction with epigenetics data

On this ongoing project, we are trying to predict HiC interaction between two regions using epigenetic signals. HiC interaction can have major roles in tissue-specific expression and how regulatory sequence variants impact complex phenotype including diseases such as cancer diabetes and obesity. There are very few HiC datasets available at high resolution since the experiment is very expensive. As a result, the computational approach for predicting long-range interaction.

In this work, we are building feature matrices from the epigenetic signals which then processes by a convolutional neural network for predicting the interaction between two regions. In this work, we will be investigating combinatorial relationships among features, and the ability of the predicted interactions to identify, topologically associated domains (TADs), loops, contact counts in different cell lines etc.

Fig: HiC interaction prediction model

Selected Publications

  1. MA Hasan, S Lonardi, "Epi2En: A Convolutional Neural Network for Genome-wide Enhancer Prediction", Submitted to BMC Genomics, 2019
  2. MA Hasan, S Lonardi "DeeplyEssential: A Deep Neural Network for Predicting Essential Genes in Microbes" In The 6th International Workshop on CNB: Modeling, Analysis, and Control, New York, 2019, to appear in BMC Bioinformatics. Student Travel Award from NSF.
  3. Stefano Lonardi, María Muñoz‐Amatriaín, Qihua Liang, Shengqiang Shu, Steve I Wanamaker, Sassoum Lo, Jaakko Tanskanen, Alan H Schulman, Tingting Zhu, Ming‐Cheng Luo, Hind Alhakami, Rachid Ounit, Abid Md Hasan, Jerome Verdier, Philip A Roberts, Jansen RP Santos, Arsenio Ndeve, Jaroslav Doležel, Jan Vrána, Samuel A Hokin, Andrew D Farmer, Steven B Cannon, Timothy J Close "The genome of cowpea (Vigna unguiculata [L.] Walp.)", The Plant Journal, 98 (5), 767-782, June 2019
  4. MA Hasan, S Lonardi "mClass: Cancer Type Classification with Somatic Point Mutation Data", Proceedings of 16th RECOMB Comparative Genomics Satellite Workshop (RECOMB-CG), pp. 131–145, Magog-Orford, Canada, 2018
  5. A. Polishko, M. A. Hasan , W. Pan, Evelien M. B. Karine L. R. Stefano L. “ThIEF: Finding Genome-wide Trajectories of Epigenetic Marks”, Proceedings of the Workshop on Algorithms in Bioinformatics (WABI), 19:1–19:16, Boston, MA, 2017
  6. Hasan M A, Hasan M K and Mottalib M A Linear Regression based Feature Selection for Microarray Data Classification, Int. J. Data Mining and Bioinformatics, 2015, Vol 11, Issue 2, pp. 167-179.
  7. K M Mutakabbir, S S Mahin, M A Hasan, Mining Frequent Pattern Within a Genetic Sequence Using Unique Pattern Indexing and Mapping Techniques, IEEE International Conference on Informatics, Electronics & Vision (ICIEV), pp. 1–5, 2014, Dhaka, Bangladesh.
  8. Hasan A, Adnan M A, High Dimensional Microarray Data Classification Using Correlation Based Feature Selection, International Conference on Biomedical Engineering (ICoBE), pp. 319 – 321, 2012, Penang, Malaysia.