arXiv subjects

Rachel Karchin

Publications and source records attributed to Rachel Karchin.

4 recordsLinked to original sources

arXiv2026

pVACtools v6: A comprehensive suite for neoantigen prediction, visualization, and therapy design

With the rise of checkpoint blockade therapies and neoantigen-based vaccines reaching later-stage trials, there is a growing need for computational tools to identify and prioritize neoantigens. pVACtools, initially introduced in 2016, is an open-source informatic suite designed to support basic and translational neoantigen research. pVACtools assists prediction, prioritization, and visualization of neoantigens, as well as design of neoantigen-based therapies. We describe several major advances to pVACtools since the last update: (1) expanded neoantigen quality and safety assessment features, including support for peptide presentation scoring, immunogenicity prediction, anchor residue analysis, reference proteome similarity, percentile score calculation; (2) addition of pVACsplice, a new tool for predicting neoantigens from tumor-specific cis-splicing mutations; (3) addition of pVACbind, a flexible tool that supports noncanonical neoantigen sources; (4) improvement in neoantigen selection strategies; (5) a substantially improved pVACvector algorithm that achieves higher DNA/mRNA vector vaccine design success rates with shorter runtimes; (6) new utilities to support synthetic long peptide vaccine design; (7) extended prediction support for many non-human species; and (8) addition of pVACcompare, a tool to support comparison between two pVACseq results. Together, these updates reinforce pVACtools as the field's most comprehensive toolkit for neoantigen research, from basic discovery to the design and execution of personalized cancer vaccine clinical trials.

q-bio.QM↗

arXiv2025

ImmunoNX: a robust bioinformatics workflow to support personalized neoantigen vaccine trials

Personalized neoantigen vaccines represent a promising immunotherapy approach that harnesses tumor-specific antigens to stimulate anti-tumor immune responses. However, the design of these vaccines requires sophisticated computational workflows to predict and prioritize neoantigen candidates from patient sequencing data, coupled with rigorous review to ensure candidate quality. While numerous computational tools exist for neoantigen prediction, to our knowledge, there are no established protocols detailing the complete process from raw sequencing data through systematic candidate selection. Here, we present ImmunoNX (Immunogenomics Neoantigen eXplorer), an end-to-end protocol for neoantigen prediction and vaccine design that has supported over 185 patients across 11 clinical trials. The workflow integrates tumor DNA/RNA and matched normal DNA sequencing data through a computational pipeline built with Workflow Definition Language (WDL) and executed via Cromwell on Google Cloud Platform. ImmunoNX employs consensus-based variant calling, in-silico HLA typing, and pVACtools for neoantigen prediction. Additionally, we describe a two-stage immunogenomics review process with prioritization of neoantigen candidates, enabled by pVACview, followed by manual assessment of variants using the Integrative Genomics Viewer (IGV). This workflow enables vaccine design in under three months. We demonstrate the protocol using the HCC1395 breast cancer cell line dataset, identifying 78 high-confidence neoantigen candidates from 322 initial predictions. Although demonstrated here for vaccine development, this workflow can be adapted for diverse neoantigen therapies and experiments. Therefore, this protocol provides the research community with a reproducible, version-controlled framework for designing personalized neoantigen vaccines, supported by detailed documentation, example datasets, and open-source code.

q-bio.GN↗

arXiv2024

Evaluation of simulation methods for tumor subclonal reconstruction

Most neoplastic tumors originate from a single cell, and their evolution can be genetically traced through lineages characterized by common alterations such as small somatic mutations (SSMs), copy number alterations (CNAs), structural variants (SVs), and aneuploidies. Due to the complexity of these alterations in most tumors and the errors introduced by sequencing protocols and calling algorithms, tumor subclonal reconstruction algorithms are necessary to recapitulate the DNA sequence composition and tumor evolution in silico. With a growing number of these algorithms available, there is a pressing need for consistent and comprehensive benchmarking, which relies on realistic tumor sequencing generated by simulation tools. Here, we examine the current simulation methods, identifying their strengths and weaknesses, and provide recommendations for their improvement. Our review also explores potential new directions for research in this area. This work aims to serve as a resource for understanding and enhancing tumor genomic simulations, contributing to the advancement of the field.

q-bio.GN↗

arXiv2009

Accumulation of driver and passenger mutations during tumor progression

Major efforts to sequence cancer genomes are now occurring throughout the world. Though the emerging data from these studies are illuminating, their reconciliation with epidemiologic and clinical observations poses a major challenge. In the current study, we provide a novel mathematical model that begins to address this challenge. We model tumors as a discrete time branching process that starts with a single driver mutation and proceeds as each new driver mutation leads to a slightly increased rate of clonal expansion. Using the model, we observe tremendous variation in the rate of tumor development - providing an understanding of the heterogeneity in tumor sizes and development times that have been observed by epidemiologists and clinicians. Furthermore, the model provides a simple formula for the number of driver mutations as a function of the total number of mutations in the tumor. Finally, when applied to recent experimental data, the model allows us to calculate, for the first time, the actual selective advantage provided by typical somatic mutations in human tumors in situ. This selective advantage is surprisingly small, 0.005 +- 0.0005, and has major implications for experimental cancer research.

q-bio.PE↗