Publications
📚 Peer-Reviewed Publications
As bioinformatics co-author contributing computational analysis expertise
🧬 Transcription factor Zfx regulates tumor’s evasion to T cell killing in immunotherapy
Journal: iScience (2025) | DOI: 10.1016/j.isci.2025.113842
Authors: Kaufmann U, Wang J, Callow M, Fortin J, Myrta S, Nickles D, et al.
Bioinformatics Contributions:
▸ CRISPR Screen Analysis: Processed ~160,000 sgRNAs using crisprVerse (screenCounter), TMM normalization on non-essential genes, limma-voom with robust empirical Bayes
▸ RNA-seq (Bulk): HTSeqGenie pipeline, GSNAP alignment (GRCh38), differential expression (edgeR, voom/limma), pathway enrichment (MSigDB Hallmark)
▸ ChIP-seq Integration: Public data analysis (GEO: GSE102616), BWA alignment, promoter enrichment (ZFX binding sites in 4 cancer cell lines)
▸ Clinical Data Analysis: TCGA integration, Cox proportional hazards models for survival, correlation analysis (ZFX expression vs patient outcomes)
▸ Visualization: Volcano plots, heatmaps, Kaplan-Meier curves, scatter plots
Key Findings: Identified ZFX as biomarker for immunotherapy response; ZFX regulates apoptosis pathway genes including CASP3
Data: GEO accessions GSE276964 (in vitro), GSE276965 (in vivo)
🧬 T cell-dependent bispecific therapy enhances innate immune activation and antibody-mediated killing
Journal: Cancer Immunology Research (2024) | DOI: 10.1158/2326-6066.CIR-23-0800
Authors: Besla R, Penuel E, Rosario GD, Cosino E, Myrta S, Polson AG, et al.
Bioinformatics Contributions:
▸ RNA-seq (Sorted NK Cells): Smart-Seq V4 library prep (2ng input), HTSeqGenie processing, edgeR (logCPM), voom/limma differential expression (baseline vs 24h TDB)
▸ Gene Set Enrichment: fgsea with MSigDB collections (c2, c5, c7), pathway visualization (IFN, TNF, interleukin pathways)
▸ Cytokine Analysis: Luminex multiplex data integration (Human CD8+ T-cell 17+plex panel)
▸ Flow Cytometry Data: FlowJo analysis support, gating strategy optimization
▸ Visualization: Heat maps, bar graphs with multiple comparisons, pathway network diagrams
Key Findings: TDB treatment activates NK cells via IL2/IL10 axis; enhanced ADCC observed in vitro and in vivo
Data: EGA accession EGAS00001006914
🧬 ERBB signalling contributes to immune evasion in KRAS-driven lung adenocarcinoma
Journal: bioRxiv (2023) | DOI: 10.1101/2023.07.24.550274
Authors: Laing S, Kruspig B, Shaw R, Officer-Jones L, Edwards S, McKinven D, Myrta S, et al.
Bioinformatics Contributions:
▸ Bulk RNA-seq: Multiple timepoints (8w, 12w, endpoint), Metacore GeneGO pathway analysis, differential expression
▸ Single-Cell RNA-seq: 10X Genomics Chromium (40,000 cells/sample), 22 clusters identified using ImmGen databrowser, PANTHER overrepresentation analysis
▸ Pathway Enrichment: EGFR/ERBB signaling (33.5x enrichment, FDR 1.06e-2), immune evasion pathways
▸ Whole Exome Sequencing: COSMIC signature analysis (APOBEC mutagenesis signatures)
▸ Visualization: UMAPs, violin plots, correlation plots (AREG vs p-ERBB2), heatmaps
Key Findings: ERBB ligand upregulation (AREG, HBEGF) mediates immune escape post-αPD-1 treatment; Afatinib restores immune infiltration
🧬 Transcriptional subtypes resolve tumor heterogeneity and identify vulnerabilities to MEK inhibition in lung adenocarcinoma
Journal: Clinical Cancer Research (2021) | DOI: 10.1158/1078-0432.CCR-20-1835
Authors: Daemen A, Cooper JE, Myrta S, Wongchenko MJ, Lin E, Long JE, Foreman O, Junttila MR, et al.
Bioinformatics Contributions:
▸ Subtype Discovery: Consensus nonnegative matrix factorization (NMF) across >800 patients (TCGA + clinical trials)
▸ Classifier Development: PAM (Prediction Analysis for Microarrays) - 113-gene signature with 87-91% cross-cohort validation
▸ Multi-Cohort Integration: TCGA, OAK trial (Phase III), POPLAR, gp28363 trials; harmonization of batch effects
▸ Gene Set Enrichment: Camera method (limma) + MSigDB Hallmark collection
▸ Drug Response Modeling: Spearman correlations (subtype score vs 526 compounds, 89 cell lines), ANOVA for predictive modeling
▸ Survival Analysis: survfit, Cox proportional hazards models, interaction terms (STK11 × subtype)
▸ Preclinical Models: 175 cell lines, 232 PDX, 108 GEMM tumors analyzed for subtype concordance
▸ Visualization: Heatmaps (z-scored, saturated ±1.5), Kaplan-Meier curves, bar graphs, volcano plots
Key Findings: Identified 3 subtypes (MUC, PRO, MES) with differential MEK inhibitor response; MES tumors benefit most from cobimetinib
Code: GitHub Repository
🧬 Inhibition of phosphodiesterase-4 promotes oligodendrocyte precursor cell differentiation and enhances CNS remyelination
Journal: EMBO Molecular Medicine (2013) | DOI: 10.1002/emmm.201303123
Authors: Syed YA, Baer A, Hofer MP, González GA, Rundle J, Myrta S, et al.
Bioinformatics Contributions:
▸ Microarray Analysis: Rat Exon 1.0ST array (Affymetrix), RMA normalization, limma (moderated t-test), FDR correction (Storey & Tibshirani)
▸ Differential Expression: OPC differentiation timecourse (0h, 4h, 12h); identified Mapk pathway regulation
▸ qRT-PCR Validation: TaqMan assays, normalization to β2-microglobulin
▸ Western Blot Quantification: Densitometry (ROD) for p-Erk1/2, p-p38Mapk, p-Creb1
▸ Statistical Analysis: GraphPad Prism (ANOVA + Dunnett’s, Mann-Whitney U, t-tests)
Key Findings: PDE4 inhibition (rolipram) accelerates OPC differentiation and remyelination via cAMP-Mapk-Creb1 signaling
Data: GEO: GSE50042
📄 Conference Abstracts
Transcriptional heterogeneity in lung adenocarcinoma reveals distinct therapeutic vulnerabilities
Conference: AACR Annual Meeting 2020
Publication: Cancer Research (2020) | Abstract PO-101
Authors: Daemen A, Cooper JE, Myrta S, Wongchenko MJ, et al.
Preliminary data presentation of LUAD subtyping project (later published in Clinical Cancer Research 2021)
🔬 Research Metrics
6 Publications
Peer-reviewed journals
3 Top-Tier Journals
Impact Factor >10
>800 Patients
Analyzed across studies
5 Public Datasets
GEO/EGA depositions
🎯 Expertise Demonstrated
RNA-seq Analysis
✓ Bulk RNA-seq (HTSeqGenie, GSNAP, edgeR, DESeq2, limma-voom)
✓ Single-cell RNA-seq (10X Genomics, Seurat-equivalent, UMAP, clustering)
✓ Smart-Seq V4 ultra-low input (2ng to 150pg cDNA to Nextera XT)
Genomic Screens
✓ CRISPR screens (~160K sgRNAs, crisprVerse, TMM normalization)
✓ ORF overexpression screens
✓ Drug response screens (526 compounds, correlation modeling)
Multi-Omics Integration
✓ RNA-seq + WGS/WES + clinical data (TCGA, cBioPortal)
✓ RNA-seq + ChIP-seq (promoter binding analysis)
✓ RNA-seq + flow cytometry (immune profiling)
✓ Spatial transcriptomics integration
Machine Learning
✓ Unsupervised: Consensus NMF (subtype discovery)
✓ Supervised: PAM classifier (113-gene signature, 87-91% accuracy)
✓ Feature engineering for drug response prediction
✓ Cross-validation design (4-fold larger validation cohort)
Statistical Rigor
✓ Survival analysis (Cox models, Kaplan-Meier, log-rank tests)
✓ Multiple testing correction (FDR, Benjamini-Hochberg, q-values)
✓ Robust methods (empirical Bayes, moderated statistics)
✓ Interaction modeling (gene × treatment, STK11 × subtype)
Reproducible Research
✓ Public data deposition (GEO: 3 datasets, EGA: 1 dataset)
✓ Code sharing (GitHub: LUAD.subtypes repository)
✓ Methods transparency (detailed Materials & Methods sections)
✓ FAIR data principles (metadata, accessibility)
📖 Full Publication List
View complete publication history and citation metrics
💼 Interested in Similar Analysis?
I can apply these same methodologies to your projects:
▸ 🧬 NGS analysis (RNA-seq, scRNA-seq, ChIP-seq, CRISPR screens)
▸ 🤖 Machine learning classifier development
▸ 📊 Multi-omics data integration
▸ 🏥 Clinical trial biomarker discovery
▸ 📈 Interactive visualization & reporting