Variant filters using segregation information improve mapping of nectar production genes in sunflower (Helianthus annuus L.)

Authors: BARSTOW, ASHLEY - North Dakota State University; McNellie, James; SMART, BRIAN - North Dakota State University; KEEPERS, KYLE - University Of Colorado; Prasifka, Jarrad; KANE, NOLAN - University Of Colorado; Hulke, Brent

Submitted to: The Plant Genome
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/12/2025
Publication Date: 5/14/2025
Citation: Barstow, A.C., Mcnellie, J.P., Smart, B.C., Keepers, K.G., Prasifka, J.R., Kane, N.C., Hulke, B.S. 2025. Variant filters using segregation information improve mapping of nectar production genes in sunflower (Helianthus annuus L.). The Plant Genome. https://doi.org/10.1002/tpg2.70042.
DOI: https://doi.org/10.1002/tpg2.70042

Interpretive Summary: In genomic research, identifying genetic markers is key to understanding complex traits, but traditional methods for filtering genetic data can sometimes miss important information. In this study, we explored a new data filtering approach for mapping genes related to nectar production in sunflower. We applied a more flexible filtering method that considers how markers are expected to segregate in breeding populations. Our previous work failed to identify an important gene previously hypothesized to be involved in nectar production, likely due to overly strict filtering. Our improved approach identified nine sunflower genes related to nectar production genes in the model species Arabidopsis thaliana, as compared to zero genes identified from the previous filtering strategy. This study highlights the value of using flexible, biologically relevant filtering methods, which can lead to better results in plant genomic studies.

Technical Abstract: Accurate variant calling is critical for identifying the genetic basis of complex traits, yet filters used in variant detection and validation may inadvertently exclude valuable genetic information. In this study, we compare a common sequencing depth filter, used to eliminate error-prone markers associated with repetitive regions, with a biologically relevant filtering approach that targets expected population level Mendelian segregation. The resulting marker sets were evaluated in the context of nectar volume QTL mapping in sunflower (Helianthus annuus L.). Our previous research failed to detect a significant interval containing a sunflower homolog of CWINV4, a strong candidate gene for nectar production (HaCWINV2). We removed a local sequencing depth filter and implemented a Chi-square goodness-of-fit test to retain markers that segregate according to expected genetic ratios. We hypothesized that this will enhance mapping resolution and capture key genetic regions previously missed. We demonstrate that biologically relevant filtering retains more significant QTL and candidate genes, including HaCWINV2, and accounted for a large amount of phenotypic variation, 48.55%. In finding nine putative homologs of Arabidopsis genes with nectary function within 2 LOD units of our QTL regions, we demonstrate that this filtering strategy, which considers biological contexts, has a higher power of plausible true variant detection than the commonly used marker depth filtering strategy.