Bikash Ranjan Samal
bikassh25@gmail.com| +32486800276 | Website | LinkedIn
Programming Skills:
• Python | R | C/C++
• Bash Scripng | SQL
• Nexlow | Git | HPC
Data science skills:
• Numpy | Pandas
• Scikit | Scipy
• Tensorow | Keras
AI skills:
• Interpetable deep learning
• CNN, GNN, LSTM
• AE, VAE, GAN
• Classical machine learning
algorithms
Bioinformac skills:
• WGS analsysis
• WBGS analysis
• RNAseq analysis
• ATACseq analysis
• Biological network analysis
• Protein sequence &
structure analysis
Scienc packages and sowares:
• RDKit
• Pysam
• SAMTools, Bedtools GATK,
Picard
• Cytoscape
Web development skills:
• HTML | CSS
• Django | Galaxy
Wet lab skills:
• General biochemical assays
• Mammalian cell culture
• DNA/RNA extracon
• Operang Tapestaon and
Varioskan
Selected Work Experience
Ghent University | Centre for Medical Biotechnology VIB-UGent Belgium
PhD student | Bioinformacian | Prof. Katleen De Preter’s Lab
Oct 2020 - Present
Interpretable AI in Cancer Drug Repurposing
• Developed interpretable deep learning models to predict drug sensivity in cancer cell lines and idenfy
potenal drug repurposing opportunies.
• Leveraged ANN architecture mirroring biological networks for enhanced interpretability.
• Employed neuron weights, neuron embeddings, and gradient analysis to reveal mulple layers of interpretability
within the model.
• Ulized transcriptomic, mutaonal and methylaon data to train robust predicve models.
• Published an article on “Opportunities and challenges in interpretable deep learning for drug sensitivity
prediction of cancer cells” (DOI: 10.3389/fbinf.2022.1036963).
Cancer Paent Cell-free DNA Whole Genome Sequence Analysis
• Developed bioinformatic pipeline for analyzing NGS data of plasma cfDNA from cancer patients to infer
diagnostic and prognostic outcomes.
• Extracted and ulized genec and epigenc features such as copy number prole, nucleosome footprints and
sequence mofs from cfDNA whole genome sequence data for predicve modelling.
• Analyzed ATAC-seq and cfRRBS (DNA methylation) data for the selection of genomic regions for predictive
modelling.
• Employed deconvolution algorithms to estimate tumoral DNA and subclonal fraction in patient blood.
Cell-free DNA characterizaon from blood sample and cell line culture
• Isolated cell-free DNA from patient / mice PDX blood samples, followed by characterization of DNA
concentration and size profiles.
• Cutured cancer cell lines in pre-condioned media, followed by characterization of DNA (from media
supernatant) concentration and size profiles.
• Isolated extracellular vesicles (EVs) from blood / cell line culture media, extracted DNA from EV lysates, and
subsequently characterized and sequenced the DNA.
MICALIS, INRAe Jouy-en-Josas | France
Bioinformatician | Prof. Jean-Loup Faulon’s Lab
Sep 2019 - Sep 2020
Deep Learning-Based Predicon of Novel Enzymac Reacons and Pathways
• Applied deep learning to train on enzymatic reactions from the BRENDA database and predict
feasibilty of denovo reactions.
• Utilized generative adversarial networks to create protein sequences tailored to specific enzymatic
reactions.
• Conducted comprehensive protein sequence analysis, including Pfam detection, calculation of
conservation score and physiochemical features for predictive modelling.
• Acknowledged for contribution in “The automated Galaxy-SynBioCAD pipeline for synthetic biology
design and engineering” (DOI: 10.1038/s41467-022-32661-x).
Repurposing and opmizing ML models used in exisng projects
• Implemented an active learning loop to enhance flavonoid production in cell-free systems.
• Acknowledged for contribution in “Large scale active-learning-guided exploration for in vitro protein
production optimization” (DOI: 10.1038/s41467-020-15798-5).
Indian Institute of Technology Kharagpur | India
Master Student | Teaching Assistant | Prof. Ranjit Prasad Bahadur’ Lab
Reconstrucon and analysis of gene regulatory network for intrinsic subtypes of breast cancer
• Devised a more accurate classification model for intrinsic subtypes of breast cancer based on their
gene expression profiles using supervised machine learning.
• Reconstructed individual gene regulatory networks for each subtype using a supervised machine
learning approach, leveraging multi-omics data including gene expression, methylation, CNV, and
miRNA expression data.
• Analyzed subtype-specific regulatory patterns to identify distinctive features and critical targets within
the gene regulatory networks.
Educaon
Indian Instute of Technology
Kharagpur
MTech in Biotechnology and Biochemical Engineering
Thesis: Reconstrucon and analysis of gene regulatory network for intrinsic subtypes of
breast cancer
2017 - 2019
Odisha University of
Technology and Research
Bhubaneswar
BTech in Biotechnology
Thesis: Isolaon and characterizaon of chinase producing bacteria
2013 - 2017
Conferences
CNAPS 13th Internaon Symposium on Circulang Nucleic Acids in Plasma and Serum | Graz, Ausa
March 2024
EMBO symposium on regulatory epigenomics 2019 | India
March 2019
EMBO Workshop | VIZBI 2019 | EMBL Hiedelberg, Germany
March 2019
Internaonal Conference on Contemporary Anmicrobial Research 2018 | IIT Kharagpur, India
Dec 2018
Awards | Scholarships | Compeons
Gandhi Young Technological Innovaon Award felicitated by the DST Minister and the Vice President of India
July 2019
GATE Fellowship by Ministry of Human Resource and Development, Govt. of India
2017 - 2019
Medhabru Scholarship for academic excellence awarded by Dept. of Higher Educaon, Govt. Of Odisha
2013 - 2017
DREAM Single Cell Transcriptomics Challenge | Among top 25 performers out of 382 parcipants worldwide
Sep - Nov 2018