Repository logo
Log In(current)
  1. Home
  2. Colleges & Schools
  3. Graduate School
  4. Masters Theses
  5. Foundations for Wild Citrus Introgression Through Structural Variant Discovery and Pangenome-Based SNP Calling
Details

Foundations for Wild Citrus Introgression Through Structural Variant Discovery and Pangenome-Based SNP Calling

Date Issued
May 1, 2026
Author(s)
Sisler, Patrick
The University of Tennessee
Advisor(s)
Margaret Staton
Additional Advisor(s)
Gautam Shirsekar
Scott Emrich
Chris Dardick
Permanent URI
https://trace.tennessee.edu/handle/20.500.14382/57801
Abstract
Crop improvement for disease tolerance and environmental resilience can benefit from accurate identification of genetic variants from distantly related or wild relatives. However, sequencing from highly diverged and distantly related species can suffer from reference bias, which limits the accuracy of variant calling using a single linear reference genome. Sequences from these species may align incorrectly or favor the reference allele, causing important variants to be missed or interpreted incorrectly. An example of how this is critically important is in the Citrus fruit industry , which isfacing a crisis due to Hunglongbing (HLB). Wild Australian lime species carry HLB tolerance and can be crossed with domesticated Citrus to genetically map and introgress this trait. Breeders are interested in the genomic basis of this tolerance, as management efforts to alleviate this disease are costly and do not offer a long lasting solution. In this work, we present comprehensive variant identification using traditional and contemporary methods using a diverse set of citrus lineages. First, using PacBio HiFi long reads, structural variants (SV) were identified across six citrus accessions representing four species using three independent callers and merged to generate a high confidence callset. The result of the consensus based analysis revealed extensive genomic rearrangements within and between the wild Australian limes, indicating potential adaptations that may be important drivers to their unique abiotic and biotic defense systems. However, these large and numerous rearrangements may be partially influenced by reference bias resulting from the use of a single citrus cultivar reference for this analysis. To address this limitation, we further characterize variation by analyzing SNPs using short read sequences from breeding program progeny, comparing between a graph-pangenomic and linear reference framework. We evaluated SNP accuracy using parent-child trios and the expected Mendelian Inheritance Error Rate (MIERMEIR) to benchmark SNP calls across a combination of six different mappers and SNP callers. The results showed that graph-based mapping improves allelic balance, and thus improves reference bias compared to a single linear reference. We also found that projecting reads to the pangenome’s backbone and calling SNPs using linear-based callers shows high similarity with linear methods while partially mitigating reference bias. Finally, we explore the cause of incorrect SNP calls, finding that pangenome graph clipping is found to be highly correlated with incorrect SNP calls.
Subjects

Biology, Bioinformati...

Citrus

SNPs

SVs

HLB

Pangenomes

Degree
Master of Science
Major
Life Sciences
Embargo Date
May 1, 2027

University Libraries

1015 Volunteer Boulevard
Knoxville, TN 37996
865-974-4351

Map & Directions
Donate to the Libraries
  • About
  • John C. Hodges Society
  • Speaking Volumes magazine
  • Outreach
  • Directory
  • Employment
  • Policies
  • Library Intranet
University of Tennessee power T logo

The University of Tennessee, Knoxville
Knoxville, Tennessee 37996
865-974-1000

Events
A-Z
Apply
Privacy
Map
Directory
Give to UT
Accessibility

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science