I gained independent research experience as an undergraduate research assistant at the Jovanovic Lab, where I also completed the Summer Undergraduate Research Fellowship (SURF), before moving on to a Research Associate role at the Broad Institute. Find a non-exhaustive summary below.

🧪 Proteomics Platform @ Broad Institute of Harvard and MIT

At the Proteomics Platform, I worked as a Research Associate (RA) for the Clinical Proteomic Tumor Analysis Consortium (CPTAC) team, generating and analyzing proteomic and post-translational modification datasets for various cancer types, including sarcoma, thyroid carcinoma, oligodendroglioma, and low-grade glioma.

At this point, you might be wondering: what does this consortium with its mouthful of a name do? CPTAC is a nationwide initiative by the National Cancer Institute focused on advancing cancer research through large-scale proteomic and genomic analyses, known as proteogenomics. Bringing cancer biologists, physicians, physician-scientists, data scientists, and computer scientists together under one roof, CPTAC is a massive undertaking in driving precision oncology through impactful discovery-based and translational studies.

Core CPTAC work

Meanwhile, (my fellow CPTAC comrades) and I

  • executed the MONTE workflow on CPTAC tumor samples, enabling multi-omic analysis of the immunopeptidome, ubiquitylome, proteome, phosphoproteome, and acetylome from single tissue samples that reached the Broad Institute
  • processed 200 papillary thyroid carcinoma patient samples (data analysis has started) and 400 pediatric low-grade glioma samples (data generation still underway)
  • optimized more efficient plate-based sections of the MONTE workflow that were initially done in individual tubes, dramatically increasing our throughput and scalability (went from being able to work with 16 samples in a day to 64 samples a day)
  • presented workflow progress and quality control metrics in monthly CPTAC working group meetings
  • prepared the data freeze for the soft-tissue sarcoma proteome acquired by data-independent acquisition (DIA)
  • conducted granular data analysis on differential extracellular-matrix protein expression in the dedifferentiated liposarcoma and mixed undifferentiated pleomorphic-myxofibrosarcoma subtypes of sarcoma (manuscript submitted)
  • implemented the AUTO-SP protocol (a sample preparation platform providing automated protocols for BCA analysis, protein digestion, and PTM enrichment) using the Opentrons Flex in collaboration with Johns Hopkins University and the Pacific Northwest National Laboratory (PNNL) to validate cross-institutional replicability of the protocol using protein identification and quantification as metrics (data analysis underway)

Side projects & collaborations

When I wasn’t busy with CPTAC (usually not the case), I

  • performed proteomic and acetylomic data generation and led analysis for collaborators (link) investigating changes in H3K27M-mutated diffuse midline glioma cell lines following treatment with EP300/CBP inhibitors or degraders
  • optimized chromatography conditions for CoAnn columns by evaluating various packing lengths for a potential lab-wide transition from PicoFrit columns

🧪 Jovanovic Lab @ Columbia University

At the Jovanovic Lab, most of my work focused on understanding and building tools that enable capturing and reconstructing protein-protein interactions (PPIs). In cells, proteins almost always act in coordination with other proteins to function and bring about particular biological/chemical effects. Therefore, efforts in identifying PPIs help reveal more information about the function and effects of such networks, not only within the context of diseases but also in non-pathological functions including cell signaling and differentiation.

This is where I learned that research was nothing but negative results in pursuit of the high for occasional wins.

SURF Project: Development of CITY-seq for High-Throughput Protein-Protein Interaction Detection

I would have loved to say “I optimized” CITY-seq, a novel high-throughput method integrating combinatorial barcoding, DNA-conjugated antibodies, and next-generation sequencing to detect protein-protein interactions with enhanced subcomplex resolution. However, in hopes of optimizing, I

  • performed DNA-protein conjugation protocols using click chemistry, implementing DNA staining methods on protein gels to verify conjugation and visually inspect the amount of unreacted DNA
  • fine-tuned a qPCR-based quality control assay to evaluate the binding specificity of DNA-conjugated antibodies, including generating covalently-bound protein complexes and their corresponding conjugated antibodies
  • presented progress and updates in lab meetings and at the SURF symposium held in Spring 2023

IP-MS Project: Optimization of IP-MS Conditions for Identification of True Interactors

I would have loved to say “I refined” interactor identification in immunoprecipitation-mass spectrometry (IP-MS) experiments by incorporating correlation analysis alongside traditional enrichment-over-control methods. However, I

  • performed multiplexed IPs to assess the quality of previously untested Santa Cruz antibodies, identifying high-performing candidates for subsequent single IP-MS experiments
  • conducted 70 single IP-MS experiments using 10 distinct antibodies at varying concentrations to evaluate the impact of antibody amount on purification efficiency and data quality
  • investigated recent findings suggesting that 1–5 µg of antibody yields comparable or higher-quality data than 10 µg while conserving resources
  • led ongoing data analysis comparing correlation analysis to traditional enrichment-based methods to assess improvements in interactor identification

AND TA-DA! A PAPER (under review)!

Optimization Project: Nanobody Titration for SPIDR RNA-Protein Interaction Technology

I engaged in efforts to incorporate nanobodies into the SPIDR protocol, a technique that utilizes split-pool barcoding and antibody-bead barcoding to detect RNA-RNA and RNA-protein interactions, while also contributing to lab-wide troubleshooting of streptavidin beads and biotinylated barcodes. Rough times, glad it got resolved. However, I did conduct many many nanobody titration experiments, determining optimal bead-nanobody and nanobody-antibody binding ratios for its future integration into the workflow.

🧪 lab techniques

DNA electrophoresis, SDS PAGE, Western Blot, Silver Staining, Dot Blot, BCA assay, qPCR, mammalian/bacterial cell culture, DNA/plasmid extraction, restriction digest test, azide click-chemistry, single/multiplexed IP-MS, HPLC, LC-MS, TMT labeling, EvoSep, Kingfisher, and Bravo

💻 computational/software

Python, R, BioRender, XCalibur, Chronos, Tunes, SpectrumMill, Spectronaut, MaxQuant

And no, I am not A.I. native yet…