Back to search

Boston University

Know Thy cells: inferring phenotype from genotype

Abstract

dc:description.abstract

These are not the cancer models you are looking for. Tumor models lead cancer research, but misidentified and/or misrepresentative tumor models mislead cancer research. Patient-Derived Cell Line (PDCL), Xenograft (PDX), and Organoid (PDO) tumor models are all used frequently by researchers for drug development and biomarker discovery in cancer. The growth and propagation of tumor models provide ample opportunity for their misidentification by contamination, mislabeling, and/or genomic evolution despite best laboratory practices. In an effort to eradicate misidentified tumor models for reproducible cancer research, and to ensure that tumor models remain predictive of their patient tumors, we developed a new classifier, BarCode, to predict the cancer type of tumors models. The BarCode classification of tumor models by cancer type ensures the credibility, reproducibility, and translation of data and research. The cancer cell most travelled. Metastatic patient tumors, rather than primary patient tumors, are responsible for 90\% of cancer deaths. Cancer diagnosis, metastasis, and treatment is dependent on the Tissue Of Origin (TOO). The question is, do metastatic tumors resemble their TOO or Tissue Of Destination (TOD)? BarCode, trained on primary TCGA adult and TARGET pediatric patient tumors, was used to determined whether metastatic patient tumors resemble their primary or metastatic cancer type. Cancer of Unknown Primary (CUP) is a rare heterogeneous clinical syndrome of metastatic cancers for which the primary tumor of origin (TOO) remains elusive. CUP comprises 3 to 5\% of total cancer diagnoses \citep{Varadhachary2014CancerSite}. CUP is aggressive and has often spread to many other tissues in the body before it is found. As the primary tumor of origin is unknown and therapies are directed against primary tumors, determining the best treatment for CUP is extremely challenging. To take the "U" out of "CUP", we used the BarCode, trained on primary patient tumors, to predict the TOO of CUPs. Detecting cancer early, when it can be cured. A single blood test that can detect and diagnose cancer type is the "holy grail" of cancer diagnostics. Circulating Tumor Cells (CTCs) are primary or metastatic tumor cells that have entered the blood as single cells. CTC analyses are considered a real-time "liquid biopsy" in cancer patients. To diagnose cancer from CTC real-time liquid biopsies, we used BarCode to determine the cancer type of CTCs.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy
Level thesis:degree_level
doctoral
Discipline thesis:degree_discipline
Bioinformatics GRS
Grantor
Boston University
Year dc:date.issued
2020

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Selby, Heather Marie
Advisor dc:contributor.advisor
  • Kon, Mark
ORCID dc:identifier.orcid

Subjects

dc:subject × 3

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2144/39862

Chain of custody

source
Harvested from
Boston University
Base URL
open.bu.edu/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Selby, Heather Marie. Know Thy cells: inferring phenotype from genotype. doctoral thesis, Boston University, 2020. https://hdl.handle.net/2144/39862