Over the past year and a half, PIRL has been working on several different therapeutic ideas for NUT carcinoma which I’ll describe in future blog posts. For now, I wanted to take stock of the current landscape of options for patients with incomplete surgical resections and make sense of when existing therapies do in fact work for a small number of patients.

This blog post is to show examples explaining the utility of this approach, provide a syntactical tutorial showing implementation examples, and describe the limitations and pitfalls I encountered while implementing it.

There is a better implementation that can deliver a blazingly fast compute time by using “map-reduce” with worker pool in the rayon package of Rust…the suggested implementation delivers a substantial improvement over the Python baseline…This approach indeed leads to a 10x improvement in Rust+Python over the pure Python implementation.

One way to speed up a slow algorithm is to move the bottleneck out of Python into a more efficient compiled language such as Rust. In this post I explore nucleotide k-mer counting as an example algorithm and evaluate the performance improvement of a few different implementations. Ultimately, my fastest Rust+Python implementation is only 2.5x faster than a pure Python equivalent.

Lipid nanoparticles (LNPs) encapsulating neoantigen-encoding mRNA hold promise for the development of more potent therapeutic cancer vaccines. As a lab focused on optimizing personalized cancer treatments, establishing that we are able to formulate and effectively utilize such particles would be a step towards initiating a clinical trial, and the first stages of this translational process often involve generating data that demonstrate efficacy in vitro.

Predicting T-cell immunogenicity, or the ability of peptides to elicit a T-cell response, is a critical element of vaccine design. This post looks at the nuance surrounding immunogenicity and how standardizing its quantification across contexts remains understudied.

In this post, we compare four software tools used to detect fusion genes from long read transcriptome sequencing data: LongGF, JAFFAL, Genion, and Aeron.

I am curious whether the NUTM1 fusion proteins can elicit T-cell responses and or even whether the fusion junction generates any MHC ligands. To start looking into these more immunological questions we first have to figure out the actual fusion protein sequences, which can be trickier than you would expect.