https://orcid.org/0000-0003-3167-8060
Figures
Abstract
Characterizing the genetic diversity and population structure can determine whether there is gene flow of the natural population of Helicoverpa armigera (Hübner) under disparate climate and habitat conditions in Northwest China. In this paper, H. armigera was genotyped in various regions of Xinjiang using Genotyping-by-Sequencing (GBS). The samples were compared using the single nucleotide polymorphism (SNP) and insertion deletion (InDel) marker data. The SNPs were used to analyze the population structure and five subgroups were obtained, which was further confirmed by principal component analysis (PCA). The phylogenetic tree identified five cluster populations of H. armigera. The average values of polymorphic information content (PIC) and genetic differentiation index (Fst) are 0.1783 and 0.1293, respectively, which are at a high level. The phylogenetic tree differentiation also indicates that the genetic diversity of cotton bollworm populations in different regions of Xinjiang is low diversity, moderate differentiation, and widespread gene flow. According to correlation analysis of the source of feeding on host plants (Bt cotton and non Bt crops) of H. armigera, seven SNPs with significant differences were obtained. The most significant SNP sequence was compared with the whole genome of H. armigera, and 10 candidate genes were screened. Whether the candidate genes function are related to Bt resistance needs further verification. This study can provide scientific basis for screening Bt resistance genes and formulating refuge strategy of H. armigera in Northwest China.
Citation: Ning Z, He J, Dong J, Li W, Li R, Gao Z, et al. (2025) Genetic diversity and population structure of the natural population of Helicoverpa armigera in Northwest China using Genotyping by Sequencing (GBS) technology. PLoS One 20(11): e0336253. https://doi.org/10.1371/journal.pone.0336253
Editor: Munir Ahmad, PMAS Arid Agriculture University: PMAS-Arid Agriculture University Rawalpindi, PAKISTAN
Received: May 8, 2025; Accepted: October 22, 2025; Published: November 6, 2025
Copyright: © 2025 Ning et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: National Natural Science Foundation of China-Xinjiang Joint Fund: 31660520: Associate Professor Wei Lu the Xinjiang Uygur Autonomous Region ‘Tianshan Yingcai’ Three Rural Backbone Talent Plan: 2023SNGGGCC032: Associate Professor Wei Lu Xinjiang Academy of Agricultural Sciences Key Cultivation Project of Scientific and Technological Innovation: xjkcpy-2020004: Associate Professor Wei Lu Key Technologies Research and Development Program: 2017YFD0201903: Associate Professor Wei Lu.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Xinjiang Uygur Autonomous Region (called Xinjiang) is located in the northwest of China, which is an important high-quality cotton production base and is the largest cotton-producing region and commercial cotton base in China [1,2]. According to the data of the Statistical Yearbook, in 2020, Xinjiang’s cotton planted area was approximately 1,864,000 hectares with a yield of 5.91 million tons, accounting for 58.82% of the national planted area and 87.33% of the national total output. The H. armigera belongs to the family Noctuidae of the order Lepidoptera [2,3], and this species has the characteristics of high omnivory, long-distance migration ability, facultative diapause and strong reproduction, which enables the H. armigera to adapt to various ecological environments [4–7]. In Xinjiang, H. armigeras are hosted by many plants in addition to cotton and also harm such crops as corn, wheat, tomato, sunflower, and peas [8]. When bollworms damage cotton, they mainly harm young tips, buds, flowers and bolls. The second and third generations of H. armigera are the most serious pests, exhibiting population densities of up to 38 per 100 plants, insect strain rates of 18 ~ 26%, and crop yields of 20 ~ 30% [9,10].
Different from the cotton region of the North China Plain, Xinjiang is characterized by “two basins lying between three mountain ranges” (the Altay Mountains, the Jungle Basin, the Tianshan Mountains, the Tarim Basin, and the Kunlun Mountains in order from north to south), the climate conditions are quite different in Southern and Northern Xinjiang. [11–14]. The main planting areas are mainly concentrated in oasis areas, which have rich water systems at the edge of the basin; therefore, these planting areas exhibit distribution patterns of isolated islands or beads [1,15]. Moreover, the cotton planting structure in Xinjiang is relatively simple, as Bt cotton accounts for 85% of cotton planted, and the natural refuge area is insufficient. The natural population of H. armigera may develop resistance to Bt under the long-term selection pressure of Bt cotton [10,16–22]. Are mountain ranges and basin deserts geographically blocking the migration of bollworms? Is there gene flow among natural populations of H. armigera in the eastern, southern and northern regions of Xinjiang? How genetically diverse are the bollworms in this region? At present, there are few answers to these questions.
Genotyping-by-sequencing (GBS) is a common simplified genome sequencing technology that is based on the principle of enzymatic digestion of genomic DNA followed by high-throughput sequencing of the sequences at both ends of the enzymatic digestion fragment [23,24]. GBS technology can flexibly adjust the tags needed to capture the restriction sites according to the research purpose to control the range of the capture sequence [23]. GBS has been widely used in the development of high-density and high-precision SNP molecular markers for crops [24,25], but there are few studies on the genetic diversity of agricultural pests. Therefore, this study on Xinjiang’s eight major cotton-producing areas, namely, Aksu (AKS), Alaer (ALR), Changji (CJ), Hami (HM), Korla (KEL), Kashgar (KS), Shache (SC), and Shawan (SW), collected natural populations of individual H. armigera, used GBS simplified genome sequencing technology to sequence their genomes, built the GBS library, performed specific-locus amplified fragment sequencing, and developed a large number of single nucleotide polymorphisms (SNPs) in H. armigera genome tags. Based on the screening of SNP marker, the genetic diversity and population structure of H. armigera in various regions of Xinjiang were analyzed. The eastern, southern, and northern populations of H. armigera in Xinjiang were clearly identified due to genetic flow among natural populations. According to the source of bollworm feeding on host plants (Bt cotton and non Bt crops), the association analysis was carried out to screen out some SNP sites with significant differences and whether they are candidate genes related to Bt resistance. The results of this study helped to elucidate which refuge strategy formulation was suitable, as well as helping to provide a scientific basis for Bt resistance monitoring and management.
Materials and methods
Sampling genomes
Forty-nine samples of H. armigera were obtained from traps in Xinjiang Aksu and the latitude and longitude are 40.523456 ° N and 80.321789 ° E (ten samples), Alaer and the latitude and longitude are 40.637896 ° N and 81.234567 ° E (five samples), Changji and the latitude and longitude are 44.003062 ° N and 87.32926 ° E (five samples), Hami and the latitude and longitude are 42.756123 ° N and 93.481357 ° E (five samples), the latitude and longitude of the sampled cotton fields in Korla are 41.682345 ° N and 86.274561 ° E (five samples), the latitude and longitude of the cotton fields sampled in Kashgar are 39.512345 ° N and 76.123456 ° E (five samples), the latitude and longitude of the cotton fields sampled in Shache are 38.576543 ° N and 77.312345 ° E (five samples), and the latitude and longitude of the cotton field sampled in Shawan are 44.212345 ° N and 85.654321 ° E (four samples). These sampled cotton fields are all planted with Bt cotton, separated into 2 mL centrifuge tubes, and stored at −20°C.
Using a Cell/Tissue Genomic DNA Extraction Kit from Hunchek Biological Technology Co., Ltd. (Beijing), genomic DNA was extracted from forty-nine H. armigera, and 1% agarose gel electrophoresis was utilized to determine the integrity of DNA, while a NanoDrop 1000C spectrophotometer (American NanoDrop, ND-1000) and Qubit 3.0 fluorometer were employed to measure the concentration and purity of the DNA samples to ensure that their quality satisfied the requirements of sequencing DNA (namely, an OD260/OD280 ratio distribution between 1.8-2.2). The concentration of the DNA was determined to be> 50 ng/μL, and total of > 2 g DNA was obtained. Samples were stored at -20°C for later use.
Enzyme digestion to build library
Genomic DNA of forty-nine H. armigera samples was sequenced using genotyping by sequencing (GBS) technology. The restriction endonucleases Mse I and Sac I were used to fully digest the whole genomes of the samples. Fragments measuring 300−400 bp were retrieved to construct the GBS library. The library was sequenced by the Paired-End 150-bp (PE150) Illumina Nova Seq sequencing platform and completed by GenoSecc Genomics Co., Ltd. The H. armigera reference genome address is Chr13: 2,663K-3,937K - Genome Data Viewer - NCBI.
Label SNP and InDel analysis
The original data obtained by sequencing were transformed into original sequencing sequences, or reads, by base calling analysis, and the reads were filtered according to the following principles: reads with ≥10% unknown nucleotides were filtered, reads consisting of >30% low-quality nucleotides were filtered, and reads that were of low quality at the 3’end were filtered. High-quality double-ended clean reads were obtained, and sequencing quality and data volume were evaluated. The comparison rate and base mass content were evaluated by comparing data to judge the validity and accuracy of the experimental process. Clean sequencing reads were compared to the reference genome using BWA software (version 0.7.15-R1140) [26], and mutant SNPs and InDels (insert missing marker) were analyzed using GATK software (version 3.7) [27] to obtain marker data.
Population genetic structure and genetic relationship analysis
The developed SNPs and InDel markers were filtered according to the criteria of completeness >0.8 and secondary allele frequency (MAF) >0.05 [28]. Based on the screened population SNPs with high consistency, the statistical software tools Structure (version 2.3.4), Plink (version v1.90p) and Mega7 (version 7.0) were used to analyze the population evolutionary tree and population structure [29–31]. Stuck software (version 1.48) [32] was used to calculate the expected heterozygosity (HE), observed heterozygosity (HO), number of inbred lines (Fis) and genetic differentiation index (Fst). Two genetic diversity indexes, the Shannon index (H) and polymorphic information content (PIC), were also calculated.
Phylogenetic tree analysis
A phylogenetic tree is a branch diagram or tree that describes the order of differentiation between populations and is used to represent the evolutionary relationship between populations. According to the PCA scatter plot, neighbor-joining methods in Mega 7 (version 7.0) software were used to construct the phylogenetic tree, and ggtree (version 1.7.10) was used for visualization.
Linkage imbalance analysis and genome-wide association analysis
In this study, VCF Tools software (version 0.1.17) was utilized to calculate the r2 value between two loci, and was employed to estimate the LD degree in the population. Gemma (version 0.98.1) software was used to perform genome-wide association analysis using a mixed linear model with the addition of genetic relationships as a random effect. GEC (version 0.2) software developed by Li et al., [33] was utilized to calculate the number of effective markers in the population, and its suggested value (1/n, where n is the number of effective markers) was employed as the threshold for association analysis.
Results
Database construction evaluation and sequencing quality evaluation
The size of the cotton-bollworm control genome in this study was 374.30 Mb, and the comparison of the control sequencing reads with the reference genome was conducted through the MEM algorithm of BWA software (version 0.7.15-R1140). The results demonstrated that the double-ended comparison efficiency of the experiment was 77.47%, which was largely normal, indicating that the library was constructed normally and without contamination. The sequencing quality value (Q) is an important index for evaluating the error rate of single-base sequencing. The higher the Q value is, the lower the error rate of single-base sequencing is. In this study, a total of 289.383 million reads were obtained from forty-nine H. armigera samples. The average Q30 (the probability of a base sequencing error is 0.001; therefore, the base’s Q30 was 94.30%, and the average GC content was 43.77%. Q30 data showed that the sequencing quality was high and that the sequencing results were reliable.
SNP and InDel marker information statistics
Reads generated by sequencing were compared to reference sequences obtained using BWA software, and mutations were detected together with samples using GATK (version 3.7) software, including SNPs and InDels. A total of 78,098 mutation loci were obtained in this study. The average genome coverage of all samples was 5.34%, and the average coverage depth was 23.78%. Among these mutations, 70,907 were SNPs, and 7,191 were InDels. Among the SNPs, 46,937 were conversion-type (A/G and C/T), while 23,970 were transversion-type (A/C, A/T, C/G and G/T), and the conversion transversion ratio (TS/TV) was 1.96. (Fig 1)
InDel (insert missing mark) length statistics (B).
Population structure analysis
Structure analysis.
Based on filtered SNPs, the population structure of forty-nine H. armigera samples was analyzed by Structure (Version 2.3.4) software (Fig 2). The results show that when K = 5, ΔK peaks, indicating that the optimal genetic group number is five. The PCA results (Fig 3) were consistent with the results of the structure analysis. The experimental materials could be divided into five subgroups. Among these subgroups, there were eleven samples from subgroup G1, primarily from Changji and Shache. There were eight samples in subgroup G2, primarily from Aksu. There were twelve samples from subgroup G3, primarily from Kashgar. There were eleven samples from subgroup G4, primarily from Changji. Seven samples were collected from subgroup G5, primarily from Shawan. The distribution frequencies of the five subgroups were 22.45%, 16.33%, 24.49%, 22.45% and 14.28%, respectively. In PCA clustering analysis, the genetic backgrounds of H. armigera samples from various regions of the 5 subpopulations were mostly similar, but there were individuals with very different genetic backgrounds in all five subpopulations, that is, G1-G5.
Phylogenetic tree analysis.
The phylogenetic tree results are shown in Fig 4. (Fig 4) The results of phylogenetic tree analysis and cluster analysis were the same and could be roughly divided into five different clusters (I, II, III, IV, and V).
Genetic diversity analysis and differentiation of H. armigera populations
Genetic diversity analysis of H. armigera.
The observed heterozygosity and expected heterozygosity of various H. armigera populations in Xinjiang ranged from 0.1974 to 0.2177 and from 0.1076 to 0.1810 with averages of 0.2089 and 0.1199, respectively. The observed heterozygosity was the lowest in the CJ19 population and the highest in the AKS20 population, while the expected heterozygosity was the lowest in the SW20 population and the highest in the ALE19 population.
The inbreeding number Fis in various H. armigera populations in Xinjiang ranged from 0.137 to 0.2842 with an average of 0.2514. The inbreeding number of the populations used in various populations was greater than 0, among which nine populations were greater than 0.2. AKS20 has the smallest Fis, while the other nine populations have more normal 0 values when Fis values deviate from equilibrium; in other words, the observed heterozygosity and the expected heterozygosity are significantly different, and the population has seriously deviated from the normal random mating mode. The Fis value of AKS20 is the closest to 0, indicating that the observed heterozygosity and the expected heterozygosity of this population are less different, and the population is close to the ideal state of free mating.
The polymorphism information content (PIC) of the 10 populations varied from 0.1721 to 0.1821 with an average value of 0.1783. The polymorphism information content (PIC) of the 10 populations varied from 0 to 0.25, and the diversity of the 10 populations was low and abundant (as shown in Table 1).
Genetic differentiation among H. armigera
Stacks software was utilized to calculate the genetic differentiation index among bollworms. Paired Fst values between 10 bollworm populations ranged from 0.1156 to 0.1486 (as shown in Table 2) with an average Fst of 0.1299. Among the 90 paired populations, the Fst values of all paired populations were less than 0.15. These results indicated that the genetic differentiation of most populations of H. armigera was moderate, which indicated that there was extensive gene exchange among natural populations of H. armigera in Xinjiang.
Chain disequilibrium analysis.
Using a statistical method to calculate the LD decline level of five subpopulations of H. armigera in eight regions in Xinjiang (Fig 5), the R2 value of G5 was the largest, and the R2 value of G3 was the smallest. The LD coefficients of the four subgroups could be rankedas G5 > G2 > G4 > G1 > G3, and the LD attenuation curves of the five subgroups were nearly parallel to the X axis. When measuring the linkage imbalance degree r2 was less than 0.33, which indicated that the LD of the population was mainly caused by recombination, while mutation had little effect on LD. Linkage disequilibrium exists in all collinearity, which indicates that the linkage groups have undergone historical reorganization. This finding may be observed because the markers among different linkage groups are independent, and the linkage imbalance among these markers is easily broken, while the markers in the same linkage group have linkage relationships; therefore, the linkage imbalance degree is higher, the similarity degree between individuals is higher, and the genetic diversity is lower. There was no significant attenuation of LD in the five subgroups, and the degree of linkage imbalance was very high.
Genome-wide association analysis.
UPLC/MS/MS was used to determine the content of gossypol (a special compound of cotton) in the body of H. armigera [34]. It can be known that H. armigera comes from Bt cotton or non Bt crops. The Manhattan diagram drawn by Gemma (version 0.98.1) based on the host plant sources (Bt cotton and non-Bt crops) on which H. armigera feeds is shown below. The number of effective markers calculated by GEC software was 7142, and the corresponding threshold of association analysis was P = 0.000145, which was 3.34 after -log10 conversion. Genome-wide association analysis was conducted on various regions of H. armigera samples, and the results are shown in Fig 6 (Fig 6).
The dotted line represents the threshold of potential genome-wide significant association: -log10 = 3.34.
Compared with the genome of H. armigera, the results were as follows there is an associated SNP on scaffold062, scaffold134, scaffold198, scaffold200, scaffold290 and scaffold304, among which the SNP on scaffold062 is the most significant, and the accession number of its source sequence was NW_018395451.1, position 734791. Ten candidate genes were screened from 600000–800000 upstream and downstream of the locus (as shown in Table 3).
Discussion
Construction of genetic maps based on GBS simplified genome technology can cover the entire genome, eliminate limits for reference genomes, and reduce the complexity of the genome [4,25]. With this technology, the data volume is low, operation is easy, and costs are low, and this method is therefore widely applied in the development of high-density, high-precision crop SNP marker research [35]. At present, GBS plays an important role in the study of the genetic diversity and population resources of agricultural pests [36]. Wosula et al. genotyped Bemisiatabaci from cassava from 8 countries in Africa and identified 6 major groups, which were further confirmed by PCA and multidimensional analysis, with significant gene exchange being detected among the groups [37]. Perepa et al. used 85 single nucleotide polymorphisms (SNPs) to genotype populations of Helicoverpa zea captured from Landisville and Rock Springs, Pennsylvania, USA, in 2002, 2005, 2016, and 2018. Genotypic samples were divided into 16 putative populations according to collection time and location. High genetic diversity within populations and low genetic differentiation between populations were observed in 2002 and 2005 [38].
Studies based on DNA genes and limited nucleotide markers have shown that the population structure of H. armigera generally lacks subgroup structure, even among populations that are geographically far apart [39–42]. Anderson et al. used genome-wide SNP markers to demonstrate differences between Australian/New Zealand bollworm subspecies and Old World bollworm subspecies but not among global populations [43]. Genotypes of H. armigera in southeastern South America and northern/central Brazil and southwestern H. armigera populations are rare and/or lack the dominant mtDNA haploid phenomenon; thus, H. armigera invasion may employ two different paths in Brazil and may be associated with the import and export of agricultural products and horticultural products [44].
Xinjiang Province covers an area of approximately 1/6 of China’s land area and is distinguished from the cotton region of the North China Plain by its unique geomorphological features of “three mountain ranges two basins” [11–14]. Unfortunately, in 2020, COVID-19 virus was rampaging around the world. Due to epidemic prevention requirements and control and travel restrictions, at some sample collection sites in Xinjiang, H. armigera samples were not collected in 2020. Among the population structures characterized in this study, the genetic diversity of the G3 subgroup was the most complex, and the samples included 6 different regions. The G1 subgroup is mainly from Changji and Shachi, but the distance between the two regions is more than 1400 km, meaning that the bollworms need to bypass the Tianshan Mountains and the Taklimakan Desert to migrate. A reasonable theory is that this spread is probably the result of natural migration. The migration of H. armigera in China usually occurs between insect overwintering and non-overwintering areas, and it usually takes 1–2 nights to migrate 200–500 km to northern non-overwintering areas [45,46]. A second theory is related to the fact that Kashgar is a special economic zone in western China. Under the influence of economic trade, agricultural products with H. armigera eggs circulate between southern and northern Xinjiang. Alternatively, this spread may be related to different pest control/management strategies. Detailed conclusions need to be proven by means of radar monitoring, morphological changes, feeding biology, flight biology and reproductive biology [46]. The genetic diversity of the G2 subgroup is relatively conservative, and the intercropping area of jujube, walnut, apple and cotton is relatively large in Aksu. It may be surmised that bollworms, as omnivorous pests, can choose among more host plants or may be related to the local characteristic fruit-cotton intercropping pattern. The G4 and G5 subpopulations were obtained from Changji and Shawan, respectively, which are in northern Xinjiang and less than 200 km apart. The genetic structures of the samples were mixed, suggesting that there might be frequent gene exchange between natural populations of H. armigera in these two locations.
In 2018, Wang Dongmei, Yang Xianming, and others used the cytochrome c oxidase subunit I (COI) gene fragment to study the population structure of cotton bollworm in Xinjiang, China. The final comparison showed 28 haplotypes, with 23 in southern Xinjiang, 5 in eastern Xinjiang, and 13 in northern Xinjiang. The average genetic differentiation index measured in all samples was 0.1321, which is at a high level, indicating that the genetic diversity of the cotton bollworm population is low diversity, moderate differentiation, and extensive gene flow [8]. After 2025, Hou, Bofeng et al. analyzed the population structure of cotton bollworms in different regions of Xinjiang. The population structure analysis showed significant gene flow in various areas of Xinjiang, but the differences between the westernmost collection sites of Kashgar and Shawan, as well as with other collection sites, were the greatest. The cotton bollworms in Shawan and Kashgar had genetic differences among each other and with other populations in Xinjiang at the whole genome level [47].
In this study, the natural population of H. armigera in Xinjiang was sequenced with the aim of developing SNPs that may be associated with Bt resistance by analyzing the genotypes of SNPs and the host plant sources (Bt cotton and non-Bt crops) that the bollworm feeds on. Due to the relatively small study sample size (n = 49), the reported Tassel analysis process for large samples [48] was not employed in the construction of the simplified genome reference sequence, and the stack analysis process [32], which is more suitable for small groups, was applied. Ten candidate genes were screened by comparing the most significant SNPs, and their functions need to be further studied.
Conclusions
The library constructed by GSB technology is pollution-free, high-quality, and the sequencing results are reliable; a large number of SNPs have been successfully screened at the whole genome level of H. armigera. The natural populations of H. armigera in Xinjiang were divided into five subgroups and five cluster groups by population structure analysis and cluster analysis, and the results were consistent. The low genetic diversity of H. armigera populations in eight regions indicated that there was extensive gene flow among them. Due to the large planting area and single planting mode of Bt cotton in Xinjiang, H. armigera has potential risk of BT resistance under the long-term selection pressure of Bt cotton. Therefore, it may be necessary to consider the whole Xinjiang Province as a spatial scale when formulating refuge strategies in the future. Seven significant SNP loci and ten candidate genes were obtained by association analysis, and their gene functions need to be further studied in subsequent experiments.
Supporting information
S1 Table. Table S1 Statistical table of sequencing data.
https://doi.org/10.1371/journal.pone.0336253.s001
(DOCX)
S2 Table. Comparison rate of sequencing data.
https://doi.org/10.1371/journal.pone.0336253.s002
(DOCX)
S3 Table. Statistical table of cotton bollworm genome coverage and coverage depth.
https://doi.org/10.1371/journal.pone.0336253.s003
(DOCX)
S6 Table. Summary of InDel annotation results.
https://doi.org/10.1371/journal.pone.0336253.s006
(DOCX)
S7 Table. Statistical table of variation quantity and density of SNP and InDel markers.
https://doi.org/10.1371/journal.pone.0336253.s007
(DOCX)
Acknowledgments
The authors thank Yu-Kun Cheng and Zhi-Yong Ni from College of Agriculture, Xinjiang Agricultural University for data analysis, and Duo Lai from Institute of Fruit Tree Research, Guangdong Academy of Agricultural Sciences for manuscript revision.
References
- 1. Lu Z-Z, Zalucki MP, Perkins LE, Wang D-Y, Wu L-L. Towards a resistance management strategy for Helicoverpa armigera in Bt-cotton in northwestern China: an assessment of potential refuge crops. J Pest Sci. 2013;86(4):695–703.
- 2. Xu Y, Wu K, Li H, Liu J, Ding R, Wang F, et al. Effects of Transgenic Bt+CpTI Cotton on Field Abundance of Non-Target Pests and Predators in Xinjiang, China. Journal of Integrative Agriculture. 2012;11(9):1493–9.
- 3. Bird LJ, Drynan LJ, Walker PW. Relative fitness and stability of resistance in a near-isogenic strain of indoxacarb resistant Helicoverpa armigera (Lepidoptera: Noctuidae). Pest Manag Sci. 2020;76(12):4077–85. pmid:32543096
- 4. Cunningham JP, Zalucki MP. Understanding heliothine (Lepidoptera: Heliothinae) pests: what is a host plant? J Econ Entomol. 2014;107(3):881–96. pmid:25026644
- 5. Downes S, Kriticos D, Parry H, Paull C, Schellhorn N, Zalucki MP. A perspective on management of Helicoverpa armigera: transgenic Bt cotton, IPM, and landscapes. Pest Manag Sci. 2017;73(3):485–92. pmid:27753247
- 6. Queiroz-Santos L, Casagrande MM, Specht A. Morphological Characterization of Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae: Heliothinae). Neotrop Entomol. 2018;47(4):517–42. pmid:29450860
- 7. Chen C, Xia Q-W, Chen Y-S, Xiao H-J, Xue F-S. Inheritance of photoperiodic control of pupal diapause in the cotton bollworm, Helicoverpa armigera (Hübner). J Insect Physiol. 2012;58(12):1582–8. pmid:23063727
- 8. Wang D, Yang X, Li H, Wuwaishi A, Ding R, Li H, et al. Genetic homogeneity between populations of cotton bollworm from Xinjiang, China. Journal of Asia-Pacific Entomology. 2018;21(1):309–15.
- 9. Wu K. Monitoring and management strategy for Helicoverpa armigera resistance to Bt cotton in China. J Invertebr Pathol. 2007;95(3):220–3. pmid:17467730
- 10. Dandan Z, Yutao X, Wenbo C, Yanhui L, Kongming W. Field monitoring of Helicoverpa armigera (Lepidoptera: Noctuidae) Cry1Ac insecticidal protein resistance in China (2005-2017). Pest Manag Sci. 2019;75(3):753–9. pmid:30101444
- 11. Wang YF, Liu P, Xia Q. Study on cooling effect of irrigation in Yangtze River Basin on extreme high temperature. J CWRH. 2024;(03):1–7.
- 12. Xu J, Chen Y, Li W, Liu Z, Wei C, Tang J. Understanding the complexity of temperature dynamics in Xinjiang, China, from multitemporal scale and spatial perspectives. ScientificWorldJournal. 2013;2013:259248. pmid:23843732
- 13. Zhu B, Yu J, Qin X, Rioual P, Zhang Y, Xiong H. Formation and evolution of sand deserts in Xinjiang, Northwest China: II. The palaeo-environmental reconstruction. J Geogr Sci. 2014;24(3):539–59.
- 14. Zhou L, Zhao C, Chen Z, Zheng S. Amplitude Tomography of Lg Waves in Xinjiang and Its Adjacent Regions. Bulletin of the Seismological Society of America. 2011;101(3):1302–14.
- 15. Lu ZZ, Baker G. Spatial and temporal dynamics ofHelicoverpa armigera(Lepidoptera, Noctuidae) in contrasting agricultural landscapes in northwestern China. International Journal of Pest Management. 2013;59(1):25–34.
- 16. Wu K-M, Lu Y-H, Feng H-Q, Jiang Y-Y, Zhao J-Z. Suppression of cotton bollworm in multiple crops in China in areas with Bt toxin-containing cotton. Science. 2008;321(5896):1676–8. pmid:18801998
- 17. Wan P, Huang Y, Wu H, Huang M, Cong S, Tabashnik BE, et al. Increased frequency of pink bollworm resistance to Bt toxin Cry1Ac in China. PLoS One. 2012;7(1):e29975. pmid:22238687
- 18. Wan P, Huang Y, Tabashnik BE, Huang M, Wu K. The halo effect: suppression of pink bollworm on non-Bt cotton by Bt cotton in China. PLoS One. 2012;7(7):e42004. pmid:22848685
- 19. Xu P, Liu Y, Graham RI, Wilson K, Wu K. Densovirus is a mutualistic symbiont of a global crop pest (Helicoverpa armigera) and protects against a baculovirus and Bt biopesticide. PLoS Pathog. 2014;10(10):e1004490. pmid:25357125
- 20. Liu C, Li Y, Gao Y, Ning C, Wu K. Cotton bollworm resistance to Bt transgenic cotton: a case analysis. Sci China Life Sci. 2010;53(8):934–41. pmid:20821292
- 21. Wang L, Ma Y, Wei W, Wan P, Liu K, Xu M, et al. Cadherin repeat 5 mutation associated with Bt resistance in a field-derived strain of pink bollworm. Sci Rep. 2020;10(1):16840. pmid:33033325
- 22. Wang L, Wang J, Ma Y, Wan P, Liu K, Cong S, et al. Transposon insertion causes cadherin mis-splicing and confers resistance to Bt cotton in pink bollworm from China. Sci Rep. 2019;9(1):7479. pmid:31097777
- 23. Kishor DS, Noh Y, Song W-H, Lee GP, Park Y, Jung J-K, et al. SNP marker assay and candidate gene identification for sex expression via genotyping-by-sequencing-based genome-wide associations (GWAS) analyses in Oriental melon (Cucumis melo L.var.makuwa). Scientia Horticulturae. 2021;276:109711.
- 24. Labate JA, Glaubitz JC, Havey MJ. Genotyping by sequencing for SNP marker development in onion. Genome. 2020;63(12):607–13. pmid:32853533
- 25. Wang N, Yuan Y, Wang H, Yu D, Liu Y, Zhang A, et al. Applications of genotyping-by-sequencing (GBS) in maize genetics and breeding. Sci Rep. 2020;10(1):16308. pmid:33004874
- 26. Wang WY, Png EL. BWA-MEM and NovoAlign were compared using human whole genome second generation sequencing data. J BNU(NS). 2021;57(03):337–44.
- 27. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010;20(9):1297–303. pmid:20644199
- 28. Sunseri F, Lupini A, Mauceri A, De Lorenzis G, Araniti F, Brancadoro L, et al. Single nucleotide polymorphism profiles reveal an admixture genetic structure of grapevine germplasm from Calabria, Italy, uncovering its key role for the diversification of cultivars in the Mediterranean Basin. Australian Journal of Grape and Wine Research. 2018;24(3):345–59.
- 29. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155(2):945–59. pmid:10835412
- 30. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007;81(3):559–75. pmid:17701901
- 31. Yu G, Smith DK, Zhu H, Guan Y, Lam TT. ggtree: an r package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods Ecol Evol. 2016;8(1):28–36.
- 32. Catchen J, Hohenlohe PA, Bassham S, Amores A, Cresko WA. Stacks: an analysis tool set for population genomics. Mol Ecol. 2013;22(11):3124–40. pmid:23701397
- 33. Li M-X, Yeung JMY, Cherny SS, Sham PC. Evaluating the effective numbers of independent tests and significant p-value thresholds in commercial genotyping arrays and public imputation reference datasets. Hum Genet. 2012;131(5):747–56. pmid:22143225
- 34. Ma H, Zhao M, Xia XM, Wang YH, Dong HZ. The effect of Bt cotton on the occurrence of cotton leaf mites and its relationship with secondary metabolites. Cotton Science. 2012;24(6):481–7.
- 35. Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, et al. A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS One. 2011;6(5):e19379. pmid:21573248
- 36. Robledo D, Palaiokostas C, Bargelloni L, Martínez P, Houston R. Applications of genotyping by sequencing in aquaculture breeding and genetics. Rev Aquac. 2018;10(3):670–82. pmid:30220910
- 37. Wosula EN, Chen W, Fei Z, Legg JP. Unravelling the Genetic Diversity among Cassava Bemisia tabaci Whiteflies Using NextRAD Sequencing. Genome Biol Evol. 2017;9(11):2958–73. pmid:29096025
- 38. Perera OP, Fescemyer HW, Fleischer SJ, Abel CA. Temporal Variation in Genetic Composition of Migratory Helicoverpa Zea in Peripheral Populations. Insects. 2020;11(8):463. pmid:32717838
- 39. Nibouche S, Buès R, Toubon J-F, Poitout S. Allozyme polymorphism in the cotton bollworm Helicoverpa armigera (Lepidoptera: Noctuidae): comparison of African and European populations. Heredity. 1998;80(4):438–45.
- 40. Zhou X, Faktor O, Applebaum SW, Coll M. Population structure of the pestiferous moth Helicoverpa armigera in the eastern Mediterranean using RAPD analysis. Heredity (Edinb). 2000;85 Pt 3:251–6. pmid:11012728
- 41. Xiao‐feng C, Sheng‐jiang T, Ren‐yi L, Ying W, Dian‐mo L. Study on the Genetic Variation of the Cotton Bollworm Helicoverpa Armigera (hübner) Populations in China*. Insect Science. 2000;7(3):243–9.
- 42. Vassal J-M, Brevault T, Achaleke J, Menozzi P. Genetic structure of the polyphagous pest Helicoverpa armigera (Lepidoptera: Noctuidae) across the Sub-Saharan cotton belt. Commun Agric Appl Biol Sci. 2008;73(3):433–7. pmid:19226783
- 43. Anderson CJ, Oakeshott JG, Tay WT, Gordon KHJ, Zwick A, Walsh TK. Hybridization and gene flow in the mega-pest lineage of moth, Helicoverpa. Proc Natl Acad Sci U S A. 2018;115(19):5034–9. pmid:29610329
- 44. Arnemann JA, Roxburgh S, Walsh T, Guedes J, Gordon K, Smagghe G, et al. Multiple incursion pathways for Helicoverpa armigera in Brazil show its genetic diversity spreading in a connected world. Sci Rep. 2019;9(1):19380. pmid:31852963
- 45. Liu Y, Fu X, Mao L, Xing Z, Wu K. Identification of host plant use of adults of a long-distance migratory insect, Mythimna separata. PLoS One. 2017;12(9):e0184116. pmid:28873457
- 46. Zhou Y, Zhao S, Wang M, Yu W, Wyckhuys KAG, Wu K. Floral Visitation Can Enhance Fitness of Helicoverpa armigera (Lepidoptera: Noctuidae) Long-Distance Migrants. J Econ Entomol. 2019;112(6):2655–62. pmid:31539425
- 47. Hou B, Cai Y, Zhang J, Oakeshott JG, Wu Y. Population genomics reveals population structure and local adaptation of the Helicoverpa armigera lineage in Xinjiang, China. Pest Manag Sci. 2025;81(8):4403–15. pmid:40191960
- 48. Schubert M, Lindgreen S, Orlando L. AdapterRemoval v2: rapid adapter trimming, identification, and read merging. BMC Res Notes. 2016;9:88. pmid:26868221