Jin-Ho Lee

• · Developed a discovery pipeline to identify tumor-specific splice junctions across large-scale cancer datasets.
• · Mapped the epitope landscape by predicting high-affinity MHC-I binding peptides from non-canonical transcripts.
• · Validated the predictive model by cross-referencing findings with high-impact transcriptomic studies.
• · Evaluated synergistic strategies for co-targeting mutation-derived and splice-derived neoantigens.

Demonstrated that alternative splicing is a viable source for high-affinity epitopes, expanding target discovery beyond traditional somatic mutations.

PythonRMapSpliceRNA-SeqTCGA DatasetsMHC-I Prediction Tools

• · Developed an in silico genotyping pipeline to extract HLA alleles directly from Whole-Exome Sequencing (WXS) data.
• · Implemented predictive modeling using ANNs (NetMHCPan) to screen missense mutations for MHC-I binding affinity.
• · Engineered a consensus methodology integrating multiple HLA typing algorithms to resolve sequence ambiguities and improve diagnostic reliability.
• · Validated the pipeline by analyzing binding specificities across HLA alleles to mitigate genotyping errors in vaccine design.

Demonstrated the feasibility of a fully computational approach to neoantigen discovery, significantly reducing costs and lead times compared to traditional PCR-based clinical assays.

PythonRHLAMinerSeq2HLANetMHCPan (Neural Networks)Shell Scripting

• · Experimental Design: Developed and executed research projects focused on DNA repair pathways and super-resolution imaging of chromatin structures.
• · Data Analysis & Modeling: Leveraged MATLAB, R, and Python to process complex imaging data and perform quantitative analysis of nano-scale structures.
• · Academic Leadership: Managed international research collaborations and mentored several Bachelor and Master's students throughout the project lifecycle.

Published research findings in multiple peer-reviewed journals and secured third-party funding for continued scientific investigations.

MATLABLinuxRPythonCOMBO-FISHSPDMDBSCAN

• · Experimental Design: Developed and executed a multi-year research project at the intersection of radiation biology and neuroscience.
• · Leadership & Mentoring: Supervised and guided students, junior researchers, and technical staff through complex laboratory workflows.
• · Scientific Communication: Secured third-party funding, established research collaborations, and published findings in peer-reviewed journals.

Successfully identified new insights into neural radiation responses and presented results at international scientific conferences.

3D Cell Culture ModelingqPCRFluorescence MicroscopyFACSBiochemistryImageJMS Office

• · Re-engineered the original SNU pipeline as a fully reproducible Snakemake 8.x workflow with per-rule Conda environments — installable without manual dependency management; run locally and on GCP GPU VMs, and executor-portable to a SLURM cluster by design (per-rule resources + Conda) — a config swap, not a rewrite.
• · Implemented tumor-vs-matched-normal junction filtering against GENCODE annotation, classifying junctions as annotated, normal_shared, or tumor_exclusive.
• · Integrated patient-specific HLA typing (OptiType) and MHC-I peptide presentation prediction (MHCflurry 2.x Class1PresentationPredictor) for end-to-end neoepitope candidate ranking.
• · Added optional TCR-pMHC ternary complex structural validation via TCRdock (AlphaFold v2 backend) on Google Cloud GPU; predicted complexes rendered as interactive Mol* 3D viewer in the report.
• · MIT-licensed, public on GitHub (Jin-HoMLee/splice-neoepitope-pipeline).

Snakemake pipeline: end-to-end neoepitope discovery from RNA-Seq FASTQ to computationally modeled TCR-pMHC complex structure — patient-specific HLA alleles, tumor-exclusive junctions.

Snakemake 8.xCondaPython 3.11HISAT2 / STAROptiTypeMHCflurry 2.xTCRdock / AlphaFold v2bedtoolsGoogle Cloud GPUMol*

• · Project Leadership: Defined project scope, managed team communication, and delivered stakeholder presentations.
• · Model Development: Designed and optimized an LSTM neural network to classify temporal sequences of ASL gestures.
• · Data Engineering: Developed a pipeline for real-time landmark extraction, feature engineering, and data cleaning from video feeds.

Delivered a functional prototype for real-time inference, enabling instant feedback for sign language learners.

PythonTensorFlow (Keras)LSTMMediaPipeOpenCVGoogle CloudScikit-LearnMLflow

• · Developed a computer vision pipeline for detecting players and tracking shuttle trajectories.
• · Implemented pose estimation models to analyze player movement patterns.
• · Extracted gameplay metrics such as rally duration and player positioning.

Functional prototype capable of generating match statistics and movement analytics.

PythonOpenCVTensorFlowMediaPipe

• · Delivered lectures on Python, machine learning, SQL, and data visualization.
• · Mentored students throughout the data science lifecycle.
• · Supervised capstone projects and technical presentations.
• · Developed and improved training materials.

Successfully supported multiple cohorts in developing industry-ready data science skills.

PythonPandasNumPyScikit-LearnTensorFlowSQLDockerdbtGoogle Cloud

• · Expanded the Python-based analytics environment in Google Cloud.
• · Migrated legacy SAS reporting workflows to Python-based pipelines.
• · Explored BI capabilities using Looker and prepared the data model.
• · Conducted stakeholder management and requirements analysis.
• · Supported AML risk analysis through ad-hoc investigations of customer relationships.

Enabled transition of legacy reporting processes into a scalable cloud-based analytics environment.

Google Cloud PlatformBigQueryPythonSASLooker

• · Conducted requirements analysis with business and analytics stakeholders.
• · Mapped data tables from legacy systems to the new cloud architecture.
• · Supported integration of over 1000 analytical processes into the cloud platform.

Enabled scalable data infrastructure for personalized marketing initiatives.

Google Cloud PlatformBigQuerySQLPython