title: Profiling pathogenicity of Bovine Meat and Milk Factors in cancer by genome and transcriptome analysis
creator: Häfele, Lisa Ruth
description: Bovine Meat and Milk factors are circular DNA sequences isolated from bovine milk and serum   samples, which have been proposed to contribute to cancer development of different cancer   types by inducing chronic inflammation in exposed tissues. While experimental analyses   indicated the presence of certain BMMF sequences in different tumor types, only specific   BMMF genomes and cancer types have been targeted in experiments so far.   For this reason, I screened multiple publicly available high-throughput sequencing data sets for   a comprehensive library of BMMF genomes using the D-ViSioN algorithm to fill this   knowledge gap by in silico analysis. With this, I managed to prove the feasibility of BMMF   detection via computational tools in RNA, WGS and single cell sequencing data and developed   processing steps to filter, normalize and characterize the BMMF signal. I screened WGS and   RNA sequencing samples of 29 and 25 different cancer cohorts of the PCAWG project, RNA   sequencing data of five cancer types provided by the TCGA project, as well as, 15 healthy tissue   cohorts derived from healthy donors included in the GTEx project. Additionally, I analyzed cell   line data of the DepMap project and a single cell data set of metastatic lung cancer.   I detected BMMF sequences on the RNA and even stronger on the DNA level in tumor and   non-tumor samples of patients with a wide range of different cancer types as well as in samples   of healthy donors. The detection of BMMF group 1 targets outnumbered the detection of   BMMF group 2, 3 and 4 targets by far both on DNA and RNA level. The comparison of BMMF   detection in a set of cancer and tissue types across five different data sets revealed the highest   percentage of BMMF positive samples for ovarian, stomach and uterine cancer in RNA   sequencing data as well as for breast, kidney, lung, pancreas, prostate and stomach cancer in   WGS data. For further subtyping of the reported BMMF hits, I defined in total 26 BMMF   subgroups spanning the four main BMMF groups. Detailed analysis of BMMF detection at   subgroup level showed that for a broad set of BMMF subgroups and a broad range of different   cancer cohorts a lower BMMF signal was found in the RNA data of tumor samples compared   to matched healthy tissue samples. These findings would indicate no increased cancer risk upon   detection of these BMMF types. On the contrary, BMMF subgroup 6 in acute myeloid leukemia   and ovarian cancer, subgroup 5 in stomach cancer, subgroup 8 in uterine cancer and subgroup   21 in acute myeloid leukemia were found to be increased in the case versus control comparison   of RNA data, which are thus candidates for investigating potential high-risk patterns. While   WGS data of early-onset prostate cancer patients exhibited a higher BMMF signal in non-tumor  samples than in tumor samples of the same patients, a kidney, lung, pancreatic and prostate   II  cancer cohort each included two or more BMMF subgroups with increased BMMF detection   in tumor samples compared to non-tumor samples of the same patients. These analyses   highlighted the importance of BMMF subgroups 1, 5, 6, 7, 10 and 21, which frequently stood   out in different data sets analyzed. In addition, I characterized the specific coverage of BMMF   reads on the respective BMMF templates for the BMMF genomes C1MI.3M.1, H1MSB.1,   C1MI.2 and C1HB.4, which showed that either the entire sequence or large parts of it are   covered by BMMF reads indicating a specific detection.  With these analyses, I identified new cancer types-of-interest as well as new target BMMF   genomes for further BMMF research. The definition and characterization of BMMF-positive   cohorts and subgroups might help to understand the pathogenic phenotype of BMMFs and to   establish BMMF detection workflows helpful in diagnostic and therapeutic setup.
date: 2025
type: Dissertation
type: info:eu-repo/semantics/doctoralThesis
type: NonPeerReviewed
format: application/pdf
identifier: https://archiv.ub.uni-heidelberg.de/volltextserver/35508/1/Thesis_2024_Lisa_H%C3%A4fele.pdf
identifier: DOI:10.11588/heidok.00035508
identifier: urn:nbn:de:bsz:16-heidok-355087
identifier:   Häfele, Lisa Ruth  (2025) Profiling pathogenicity of Bovine Meat and Milk Factors in cancer by genome and transcriptome analysis.  [Dissertation]     
relation: https://archiv.ub.uni-heidelberg.de/volltextserver/35508/
rights: info:eu-repo/semantics/openAccess
rights: http://archiv.ub.uni-heidelberg.de/volltextserver/help/license_urhg.html
language: eng