Characterizing Unmapped Reads To Decipher Microbiome Lead To Identification Of Possible Dysbiosis In Fatal Graft-Versus-Host Disease Of The Intestine

Shabaneh, Ashraf
2020-04

View/Download File

Shabaneh_umn_0130M_21135.pdf (8.1 MB)

Persistent link to this item

http://hdl.handle.net/11299/215028
Statistics
View Statistics

Journal Title

Journal ISSN

Volume Title

Title

Characterizing Unmapped Reads To Decipher Microbiome Lead To Identification Of Possible Dysbiosis In Fatal Graft-Versus-Host Disease Of The Intestine

Authors

Published Date

2020-04

Publisher

Type

Thesis or Dissertation

Abstract

Acute graft-versus-host disease (GVHD) of the intestinal tract is a potentially fatal complication of allogeneic hematopoetic cell transplantation (HCT). Approximately one-half of patients do not improve with standard high-dose corticosteroids. The mechanisms underlying poor steroid response are not understood, although recent data suggests that an inbalance between commensal and pathogenic microbes may play a role in the disease. The potential role of dysbiosis in fatal acute GVHD has been studied through the lens of 16S rRNA sequencing on fecal samples but not directly at the host/microbial interface. To overcome the limitations of fecal 16S rRNA sequencing in studying GVHD microbiota, we developed a pipeline that deciphers whole transcriptome unmapped reads that establishes both host (mucosal and submucosal tissue), donor (hematopoietic cells), microbial profiles for patients with GVHD. We sequenced rectosigmoid biopsies of patients with life-threatening GVHD to show how their cellular/microbial communities shift as their stage of GVHD progresses from early stage GVHD to steroid refractory GVHD (paired sample analysis). For analysis of microbial communities, we isolated unmapped reads using feature functions from the hisat2 alignment algorithm. We aligned the unmapped reads to Pathseq’s microbial reference database, and created a microbiome reads matrix. For analysis of the hematopoietic compartment, we used the CIBERSORT cell deconvolution algorithm to establish immunology profiles of the samples. The statistical analysis we used to identify critical changes included diversity and clustering analysis with the application of unsupervised machine learning methods, along with test statistics to identify specific class defining variables. We show that profiling unmapped transcriptome sequencing reads can provide a glimpse of holistic biological profiles of patients with GVHD. This dual mucosal/microbial analysis method may be useful to further characterize dysbiosis in a wide range of diseases.

Keywords

Description

University of Minnesota M.S. thesis. April 2020. Major: Health Informatics. Advisor: Jinhua Wang. 1 computer file (PDF); x, 76 pages.

Related to

Replaces

License

Collections

Master's Theses (Plan A and Professional Engineering Design Projects)

Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Suggested citation

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.