Machine learning and analysis of genomic diversity of “Candidatus Liberibacter asiaticus” strains from 20 citrus production states in Mexico

Authors: HUANG, J - South China Agricultural Univerisity; ALANIS-MARTINEZ, L - National Station Of Plant Epidemiology, Quarantine And Sanitation (ENE CUSAV); KUMAGAI, L - California Department Of Food And Agriculture; DAI, Z - South China Agricultural Univerisity; ZHENG, Z - South China Agricultural Univerisity; Perez De Leon, Adalberto - Beto; Chen, Jianchi; DENG, X - South China Agricultural Univerisity

Submitted to: Frontiers in Plant Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/23/2022
Publication Date: 12/15/2022
Citation: Huang, J., Alanis-Martinez, L., Kumagai, L., Dai, Z., Zheng, Z., Perez de Leon, A.A., Chen, J., Deng, X. 2022. Machine learning and analysis of genomic diversity of “Candidatus Liberibacter asiaticus” strains from 20 citrus production states in Mexico. Frontiers in Plant Science. 13. Article 1052680. https://doi.org/10.3389/fpls.2022.1052680.
DOI: https://doi.org/10.3389/fpls.2022.1052680

Interpretive Summary: Huanglongbing (HLB), also known as yellow shoot disease and citrus greening disease, is a highly destructive citrus disease worldwide that threatens citrus production in North America. Understanding aspects of HLB epidemiology in the United States and Mexico is required to mitigate its economic impact in California, where it was detected in 2012 in residential citrus plants. HLB was first reported in Mexico in 2009 and is now found in multiple states. The diversity of the bacterial pathogen that causes HLB in Mexico and in the United States remains to be fully determined. In this study, 40 samples of insect vectors carrying the HLB pathogen were collected from 20 states in Mexico and used in genome sequence analysis that classified the pathogen into two groups. One group was similar to strains found in Florida and the other group was similar to strains previously detected in California. The grouping conclusion was strengthened through the application of machine learning, which was used to analyze the whole genome sequences of pathogen strains. Results from this study will benefit current HLB research and management in both Mexico and United States.

Technical Abstract: Huanglongbing (HLB) is a highly destructive citrus disease associated with a nonculturable bacterium, ‘Candidatus Liberibacter asiaticus’ (CLas), which is transmitted by Asian citrus psyllid (ACP), Diaphorina citri. In Mexico, HLB was first reported in Tizimin, Yucatán, in 2009 and is now found in multiple states. Understanding the population diversity of CLas is critical for HLB management. In this study, 40 CLas-infected ACP samples from 20 states in Mexico were collected. The CLas was detected by 16S rRNA gene (rrs)-based PCR and confirmed by nucleotide reductase gene (nrdB)-based PCR. A terL-based typing system (TTS) classified the Mexican strains into two genotypes: Term-G that included four strains from Yucatán and Chiapas, similar to strain Psy62 from Florida, USA, and Term-A that included all other 36 Mexican strains, similar to strain AHCA1 from California, USA. A Term-G strain (YTMX) and a Term-A strain (BJSMX) were sequenced using next-generation sequencing approach (Illumina Hiseq). The YTMX genome was 1,229,040 bp long and had three phages (Type 1, Type 2, and CLasMV1). The BJSMX genome had 1,230,223 bp with one Type 1 phage. The TTS was further validated utilizing CLas whole genome sequences available in GenBank database utilizing two machine learning approaches: 1) principal component analysis on average nucleotide identities with whole CLas genome in consideration and 2) sparse partial least squares regression on single nucleotide polymorphisms focusing on chromosomal genes. Collectively, genome-based analyses revealed close relationships of CLas strains between Mexico and United States and suggested at least two separate introductions of CLas in Mexico.