Projects:

• RNA-Seq analysis and Tandem Mass Tag Proteomincs analysis:
  
  • Developing and maintenaning computational pipelines using publicly available and in-house developed tools for the analysis of experiment data.
  • Monitoring statistical aspects of study/experiment design, identification of appropriate tests and models, data analysis and visualization, assuring quality of data.
  • Worked with wet lab biologists helping in data analysis and visualizations.

Projects:

• Machine learning models deconvolution:
  
  • Assessed performance of machine learning models for tumor micro-environment deconvolution using statistical methods.

Projects:

• Automated flow cytometry pipeline:
  
  • Led a project to implement an innovative automated flow cytometry data analysis pipeline using gradient boosting models, significantly reducing (1400-fold speedup) data analysis time for wet bench researchers.
  • By automating the data processing and classification steps, the pipeline improved workflow efficiency and allowed for quicker access to critical insights.
  • This work was used for multiple projects within the laboratory.
• Differential gene expression methods benchmarking:
  
  • Worked with computational biology group from the university of Warsaw on a large collaborative project focusing on the analysis of single cell RNA-seq data from hundreds of patients.
  • Benchmarked traditional approaches for differential gene expression methods vs deep learning-based approaches.
  • The results of this work were essential for downstream tasks in this project.
  • The project is still in revision-stage (2025)
• Single cell spatial transcriptomics analysis:
  
  • Researched spatial transcriptomics platforms (Nanostring CosMx, 10x Xenium) that helped bringing these technologies to the laboratory.
  • Utilized various best-practice computational tools and techniques to decipher spatial patterns and identify potential biomarkers associated with pulmonary diseases, assessed publicly available datasets.
• Single cell RNA-seq data analysis:
  
  • Analyzed single-cell data from pulmonary samples to gain insights into cellular heterogeneity and molecular changes in disease conditions.
  • Employed bioinformatics tools to preprocess and interpret single-cell RNA sequencing data, revealing critical cellular subpopulations and their functional roles.

Projects:

• Cell type deconvolution of cancer- associated fibroblasts in bulk RNA-seq data using machine learning methods with the goal to improve outcome prediction in cancer patients:
  
  • Led development of a project classifying fibroblasts in tumor microenvironments using machine learning and gene expression signatures, which resulted in my thesis.
  • Presented work at OpenBio 2022.
  • Work on this project was continued and results were submitted to AACR 2024.
• Machine learning models validation:
  Developed a script to validate machine learning models that deconvolve sample cells’ percentages based on sequencing data.
• Enhanced cross-team communication:
  Developed standard report form of RNA samples analysis to share results efficiently across multiple bioinformatics teams.
• Reproducible programming and RNA-seq analysis:
  Performed code debugging, analysis, and filtration of RNA-seq lab samples.