I am a consultant for research in pharmaceutical translational medicine, with a heavy emphasis on genomics and bioinformatics. I'm cross-trained in immunology, oncology, bioinformatics, genomics, and chemistry. My training and experience have made able to integrate information and synthesize hypotheses and conclusions across diverse domains of science and medicine.

Below is a small collection of tools I've created to answer questions about cancer and immunology genomic studies, including the analysis of single-cell RNA information.

T2T: the new version of T2, a browser for the 2018 release of TCGA (pan-cancer study group), now with Thanos interactive filtering
T2T captures and plots RNA, gene-level mutation events (.mut), position-specific mutations (.fmut), continuous and thresholded copy number estimates (.cnv and .cnc), signature projections, subtype classifications, purity estimates, survival, and much more, with advanced capabilities for adjusting for biases and covariates in multi-modal TCGA data.

New in T2T is Thanos, a cross-filtering tool for choosing exactly which samples are plotted. Pick any set of variables (expression of a gene, a mutation, copy number, a signature, a clinical annotation) and each one gets a live histogram with a slider or checkboxes. Every histogram shows the samples that pass all of the other filters, with its own selection highlighted, so as you move one filter you watch its effect on every other variable. The variables you plot appear there automatically, any other variable can be added, and the samples that survive are the ones that are plotted.

T2T also gives full control of a plot's appearance, with every ggplot setting searchable by name, and makes publication-quality figures at an exact size and resolution (PNG, TIFF or PDF) with a live preview of the finished figure.

The older version is still available: T2 (previous version).

"Faux Flow" viewer for single cell Melanoma RNA-seq data
There's only so much you can get out of a T-SNE or UMAP clustering. I built tools that are modeled on the behavior of flow cytometry tools. Here's one that interrogates the Tirosh et al. Melanoma paper (fantastic paper),

Faux Flow viewer for single cell, blood derived, monocytes and dendritic cells
Alexandra-Chloé Villani et al.

Results browser to support the 2017 Siemers et al. genome-wide association study of tumor genetics with immune cell estimates.
(Link to paper)

Partial correlations: a "cell of origin" data mining tool
In an environment of highly correlated data, where are the real correlations? This tool uses partial correlation theory to divine the cellular source of gene expression in tumors, despite the overall correlation of immune infiltrates (UCSC Toil RNA-seq).

Network relationships of genes across TCGA
The PLOS paper (referenced above) used signatures for immune cell types that were derived from a mutual-rank distance network across TCGA. In particular, we discovered the association of CCR8 with tumor Tregs using these networks. The AllNets application will let you interrogate these relationships across all of TCGA, or any specific tumor type.