I have a strong interest in Machine Learning, Deep Learning, and Bioinformatics. I am currently pursuing a B.Tech in Biotechnology at VIT, Vellore, where I am actively developing my computational and bioinformatics skills.

Feel free to reach out if you’d like to discuss projects, research, or opportunities!
LinkedIn

Email: amirtesh21.5@gmail.com
Porfolio website: Amirtesh-portfolio

• Proficient in Scikit-learn, XGBoost, LightGBM, and CatBoost for building predictive models.
• Proficient in implementing Deep Learning techniques using PyTorch for tabular and image data, with experience in building neural networks.
• Basics of TensorFlow for deep learning tasks.
• Knowledge of data manipulation and visualization libraries, including:
  • NumPy and Pandas for data manipulation and analysis.
  • Matplotlib and Seaborn for creating insightful visualizations.
• Developed a library called Torchify to simplify PyTorch workflows with features like model.compile(), model.fit(), model.predict(), and model.performance().

• Basics of data analysis and visualization using dplyr and ggplot2.
• Basics of machine learning workflows in R.

• Performing Protein-Ligand Docking using tools like AutoDock Vina, Smina, QVina, SwissDock, AutoDock 4.2, and ADFR suite for both rigid and flexible docking.
• Performing Molecular Dynamics Simulations with GROMACS.
• Performing MMPBSA analysis using gmx_MMPBSA to calculate binding free energies.
• Using molecular visualization and modification tools such as PyMol, Chimera, Python Molecular Viewer (PMV), AutoDock Tools, and Discovery Studio Visualizer.
• Usage of RDKit in Python for cheminformatics, including molecular descriptor calculation, fingerprint generation, and chemical property prediction.
• Utilizing computational tools for drug discovery, including SwissADME, Swiss Target Prediction, Swiss Similarity, Protox, MolSoft, and pkcsm.

• Usage of BioPython for bioinformatics tasks, including:
  • Reading and processing sequence files in formats like FASTA and GenBank.
  • Creating and manipulating SeqRecord and SeqFeature objects.
  • Performing sequence alignment using tools like ClustalW, MUSCLE, and EMBOSS Needle/Water (both within Python and via EMBL web tools).
  • Using BLAST for sequence similarity searches - blastn, blastp, blastx, and tblastn.
  • Working with phylogenetic data using Bio.Phylo.
  • Writing output files in various biological formats.
  • Familiar with pairwise sequence alignment for similarity studies.

• Performing quality checks on FASTQ files using fastqc.
• Trimming raw reads with trimmomatic.
• Aligning sequences to reference genomes using HISAT2.
• Building a feature count matrix using the featureCounts tool from Subread.
• Automating RNA-seq workflows for multiple FASTQ files to create combined counts data and metadata using bash scripts.
• Performing Differential Expression Analysis using DESeq2 and edgeR in R and using PyDESeq2 and DEGA in Python, including:
  • Loading counts and metadata to create DESeq datasets.
  • Performing quality control on input data (e.g., row/column name checks and removing low-count reads).
  • Running differential expression analysis and saving results.
  • Interpreting p-values and adjusted p-values (padj) and identifying significantly expressed genes.
  • Filtering results based on cutoff values (e.g., alpha and fold change).
  • Converting ENSEMBL IDs to gene names for better biological interpretation.
  • Performing quality checks on the data after differential expression analysis by building PCA plots, estimating size factors, and building dispersion plots.
• Visualizing RNA-seq results by creating MA plots, volcano plots, and heatmaps in R.
• Performing Gene Ontology (GO) and KEGG pathway analysis of top genes after differential expression analysis using the DAVID tool online.
• Performing Geno Ontology (GO) and Gene Set Enrichment Analysis (GSEA) analysis using clusterProfiler package in R.

• Epitope Prediction: Prediction of B-cell, Tc-cell, and Th-cell epitopes using IEDB (all 6 methods), NetCTL, and IEDB Class I/II.
• Epitope Property Analysis: Proficient in evaluating antigenicity (ANTIGENpro, VaxiJen), allergenicity (AllerCatPro, AllerTOP, AllerFP), and toxicity (ToxinPred) of predicted epitopes.
• Vaccine Construct Design: Constructing primary vaccine sequences by fusing epitopes with appropriate linkers and adjuvants.
• Physicochemical Property Analysis: Experienced in analyzing molecular weight, pI, GRAVY, instability index (ProtParam), and solubility (SolPro).
• Structural Predictions:
  • Secondary structure prediction using SOPMA and PSIPRED.
  • Tertiary structure modeling using Swiss-Model, I-TASSER, and AlphaFold.
  • Structural validation through Ramachandran plots and Z score analysis.
• Molecular Docking:
  • Protein-protein (receptor-vaccine) docking using HADDOCK.
• Molecular Dynamics Simulation: Experienced in performing MD simulations of vaccine-receptor complexes using GROMACS to assess stability and interactions.

• Preprocessing & Quality Control: Filtering low-quality cells, removing doublets using DoubletFinder.
• Normalization & Scaling: Using default methods in Seurat and using SCTransform for faster and more accurate normalization.
• Dimensionality Reduction & Clustering: Performing PCA, UMAP, t-SNE for visualizing cell populations.
• Differential Gene Expression Analysis: Identifying marker genes for each cluster.
• Cell Type Annotation: Assigning biological identities to clusters using marker genes.
• Trajectory Analysis: Using Monocle3 with Seurat objects to infer cell differentiation pathways and pseudotime trajectories.
• Chromatin Accessibility Analysis (scATAC-Seq): Using Signac to analyze open chromatin regions and integrate chromatin accessibility with gene expression.

• Gene Co-expression Analysis: Identifying highly co-expressed gene modules from RNA-Seq data.
• Soft Thresholding Power Selection: Using scale-free topology model fit to determine optimal soft threshold.
• Module Detection: Constructing signed networks and identifying gene modules using hierarchical clustering.
• Trait Correlation Analysis: Associating gene modules with phenotypic traits (e.g., disease severity).
• Hub Gene Identification: Detecting key hub genes in significant modules for further functional analysis.
• Visualization & Interpretation: Heatmaps, dendrograms, and module-trait correlation plots for better insights.

• Quality Control & Trimming: Using FastQC and MultiQC for QC, and Trimmomatic/Fastp for adapter trimming.
• Read Alignment: Mapping reads to a reference genome using HISAT2 (whole genome) and BWA (specific chromosome).
• BAM File Processing: Sorting, fixing coordinates, marking/removing duplicates, and indexing BAM files using Samtools.
• Variant Calling & Filtering: Identifying SNPs and Indels using FreeBayes, and filtering variants with Bcftools.
• Variant Visualization: Using plotVCF in R for visualizing called variants.

• Data Acquisition & Preprocessing: Downloading ChIP-Seq data in SRA format and converting to FASTQ using fastq-dump or fasterq-dump from the NCBI toolkit.
• Quality Control & Trimming: Performing QC using FastQC, and trimming reads with Fastp.
• Genome Alignment: Mapping reads to a reference genome using BWA.
• BAM File Processing: Sorting BAM files, filtering reads with a Q value using Samtools.
• Peak Calling: Identifying enriched regions using MACS2.
• Peak Annotation & Motif Finding:
  • Using HOMER suite (annotatePeaks.pl, findMotifsGenome.pl) to analyze regulatory elements.
  • Using ChIPSeeker (R) for peak annotation and performing Gene Ontology (GO) and KEGG pathway analysis on identified genes.

• Short-Read Assembly:

  • Quality Control & Trimming: Using FastQC, MultiQC, and Porechop for adapter trimming.
  • Genome Assembly: Assembling short reads using SPAdes.
  • Visualization: Inspecting assembly graphs using Bandage.

• Long-Read Assembly:

  • Quality Control & Trimming: Using FastQC and Porechop for long-read preprocessing.
  • Genome Assembly: Assembling long reads using Flye.
  • Visualization: Exploring assembly graphs with Bandage.

• Quality Control & Trimming: Performing QC using FastQC and trimming reads with appropriate tools.
• Taxonomic Classification: Using Kraken2 with appropriate reference databases for microbial identification.
• Abundance Estimation: Quantifying microbial abundance using Bracken.
• Visualization: Creating taxonomy-based visualizations of identified microbes using ktImportTaxonomy.

• Data Preparation & Conversion: Converting input data into PLINK binary format (BED, BIM, FAM) for efficient computation.
• Quality Control (QC):
  • Filtering samples and SNPs based on missingness thresholds.
  • Filtering based on Minor Allele Frequency (MAF) and Hardy-Weinberg Equilibrium (HWE) p-values.
• Association Testing:
  • Performing association tests using plink --assoc, --logistic, --linear, and --model commands depending on trait type (binary/quantitative).
• Visualization:
  • Creating Manhattan plots and QQ plots in R for GWAS result interpretation.
• Result Interpretation:
  • Extracting and filtering top hits based on p-value thresholds.
  • Annotating significant SNPs to genes for biological interpretation.

• Trimming: Using Trim Galore! for adapter and quality trimming of bisulfite sequencing data.
• Alignment & Methylation Extraction:
  • Bismark for genome preparation, alignment of bisulfite-treated reads, deduplication, and methylation extraction.
• Downstream Differential Methylation Analysis:
  • methylKit to identify significantly differentially methylated cytosines (DMCs) across chromosomes.
  • annotatr for annotation of DMCs with genomic features (promoters, CpG islands, etc.).
  • clusterProfiler for GO and KEGG pathway enrichment analysis on annotated genes.

• Somatic Mutation Analysis:

  • Visualizing mutation distributions and identifying key mutations using oncoplot, lollipopPlot, and rainfallPlot.
  • Exploring oncogenic drivers with Oncodrive.

• Tumor Heterogeneity and Clonality:

  • Identifying clonal vs. subclonal mutations to understand tumor evolution.
  • Performing VAF-based clustering to determine clonal populations within tumors.

• Mutation Signature Analysis:

  • Comparing COSMIC mutation signatures with own data to understand mutagenic processes.

• Co-occurrence and Mutual Exclusivity:

  • Exploring co-occurring and mutually exclusive mutations to understand their role in tumor development.