Out of the 1635 recombinant clones screened, 1309 (80%) clones had the insert of desired size (≥500 bp). Among these, 170 (13%) clones had repeats of varying lengths. From these, 90 microsatellite loci were selected and primers were designed using Primer 3 software (http://frodo.wi.mit.edu/cgi-bin/primer3) [9]. Of the 90 loci studied, the majority contained dinucleotide repeat motifs (72%), followed by tri- (15.3%), and more complex (12.7%) repeat motifs. Dinucleotide repeats mostly consisted of GT (17.7%) repeats, followed by GA repeats (12.2%). The higher percentage of dinucleotide repeats found was in contrast to the report of SSRs found in another midge, the Hessian fly [10], which had an abundance of trinucleotide repeats.

After standardization, 15 loci (Table 1) were found to be polymorphic between the biotypes and sexes, whereas 75 loci were found monomorphic (Supplementary Table 1). Number of alleles of these 15 markers ranged from two (Oosat21, Oosat46 and Oosat78) to 11 (Oosat59) per locus. All the loci had expected heterozygosity of more than 0.5 in different biotypes and therefore these markers can be considered as good for parentage analysis. The range of observed and expected heterozygosity was calculated to be 0 to 1 and 0.304 to 0.910, respectively. Polymorphic information content (PIC), a measure of informativeness of a marker, calculated using Cervus v2.0 [11], ranged from 0.305 to 0.877; 14 markers with >0.5 PIC are considered highly informative in terms of their suitability for diversity analysis. Oosat26, Oosat55 and Oosat83 were most informative, while Oosat21 the least, against all the three biotypes. Despite multiple alleles, frequency of alleles varied in different biotypes (Table 2). Allele frequency of the predominat allele for each of the biotypes varied from 0.15 to 0.8. Some markers showed high allele frequency for some alleles, therefore fixation tendency of all markers was analysed. Fixation table (Table 3) was generated with following fixation values; FIS, to measure the deviation of genotypic frequencies from panmictic frequencies in terms of heterozygous deficiency or excess, FST, reduction in heterozygosity in a subpopulation due to genetic drift and FIT, overall inbreeding coefficient of an individual relative to the total population. For three loci (Oosat24, Oosat35 and Oosat46) negative FIS values indicated heterozygote excess (outbreeding). For other locus, positive values indicated heterozygote deficiency (inbreeding). FST values up to 0.05 indicate negligible genetic differentiation within the population which was observed for locus Oosat21, Oosat83 and Oosat88. For all the polymorphic loci, the Ewens-Watterson Test for Neutrality was also performed and it was observed that except for the locus Oosat88, all the loci were found neutral. Oosat88 was not behaving neutral in GMB4M population (at 95% confidence level).

Linkage disequilibrium studies (using Fisher’s method) after sequential Bonferroni correction revealed that Oosat26, Oosat36 and Oosat88 were significantly associated with each other. Also the markers Oosat24 and Oosat46, Oosat35 and Oosat21 were associated with each other, whereas other markers segregated independently in the population. P-value (HWE) mentioned in Table 1, was calculated according to Guo and Thompson [12]. Observed deviation from Hardy Weinberg equilibrium for some markers (Oosat3, Oosat21, Oossat35, Oosat43, Oosat79 and Oosat88; P > 0.05) may be the result of inbreeding, natural selection or genetic drift. Null alleles were also detected using Microchecker v.2.2.3 [13] for markers Oosat16, Oosat26, Oosat36, and Oosat83, and these could be as a result of mutations occurring in the flanking regions, preventing one or both of the primers from binding. Other polymorphic markers did not show null alleles. Markers without null alleles, give a better estimate of allele frequencies.

BLASTX analysis was performed using the sequence information of the 15 polymorphic loci which revealed that Oosat78 showed homology to tyrosyl-dna phosphodiesterase of Culex quinquefasciatus (E value: 1 × 10⁻¹⁰). The SSR repeat sequence was found to be present in the intron region of the gene. The gene has been reported to be from the phopholipase D family, which includes diverse groups of enzymes involved in phopholipid metabolism, a bacterial toxin, viral envelop proteins, and bacterial nucleases [14]. BLASTN analysis, and also BLASTX analysis with Hessian fly sequences showed insignificant similarity.

The inheritance pattern of three of the labeled markers, Oosat43, Oosat55 and Oosat59, was studied in F₂ families of GMB4M through pedigreed crosses. Inheritance of two of the markers—Oosat55 and Oosat59—proved to be sex linked. The male progeny inherited alleles only from the female parent (Figures 1 and 2); but female progeny inherited alleles from both the male and female parents. However, the pattern of inheritance of alleles of Oosat43 suggested an autosomal pattern as the alleles were inherited from both the male and female parents (Figure 3). Inheritance pattern of these markers confirmed existence of sexual dimorphism [15] and the abnormal chromosomal cycle typical of Cecidomyiidae [16,17]. Sex linked markers may be useful in tagging virulence alleles that are often sex linked [18]. Further, studies on inheritance of the markers also suggested some degree of instability with the appearance of novel non-parental alleles appearing in the offspring (Figures 1 and 2). Thus, the present study reports and confirms the usefulness of SSR markers for the rice gall midge.

Colonies of gall midge biotypes (GMB) GMB1, GMB4 and GMB4M are being maintained in the greenhouse at the Directorate of Rice Research under physical isolation and on appropriate differential rice genotypes [5]. Iso-female families were initiated with two founding pairs, since a single female will produce either all male (androgenic) or all female (gynogenic) progeny. Eight to ten F₁ pairs were mated separately to obtain F₂ adults, which were then pooled to initiate the family. At least 10 generations were reared before using these insects for DNA extraction and for inheritance studies.

DNA was extracted from the iso-female families of adult midges. The insects were crushed in an extraction buffer (0.1 M NaCl; 0.1 M Tris-HCl, pH 9.1; 0.05 M EDTA; 0.05% SDS), and extracted once with phenol:chloroform:isoamyl alcohol (25:24:1), and then once with chloroform:isoamyl alcohol (24:1). The purified genomic DNA was ethanol precipitated and resuspended in sterile distilled water after rinsing the pellet in 75% alcohol. The pooled DNA from the three biotypes was used for the purpose of library generation.

The library was constructed using hybridization capture approach of Glenn and Schable [19]. Genomic DNA was digested with the restriction enzymes Rsa I and Xmn I and ligated with the super SNX double stranded linkers on both sides to provide the primer binding site for subsequent PCR steps. They also provided sites for cloning in the vectors. Dynabead enrichment for Microsatellite containing DNA was performed. From the hybridized DNA + probe mixture, the DNA fragments with microsatellite repeat were captured using Dynabeads (Dynal, Oslo, Norway) under magnetic field. The amount of eluted DNA was increased by using the PCR enrichment to recover enriched DNA fragments using the super SNX forward primer with the appropriate PCR components under the conditions: 95 °C for 2 min; then, 25 cycles of 95 °C for 20 s, 60 °C for 20 s, 72 °C for 1.5 min; then 72 °C for 30 min. The enriched PCR product was directly cloned into TA vector (Invitrogen, U.S.). Plasmids were isolated and colony PCR was performed. The plasmids containing insert sizes of above 500 bp were selected and diluted to 100 ng/μL, and sequenced using automated sequencer (ABI prism 3700). The sequences were screened for the presence of microsatellites using the software MICAS (www.cdfd.org.in/micas/) after removal of redundant sequences. Primers were designed from the flanking region of the repeats of the non-redundant sequences using Primer3 software [9].

DNA was isolated individually from 10 female and 10 male adults from the iso-female families each of GMB1, GMB4 and GMB4M by Hot Shot protocol [20]. These DNA samples were used as templates directly for PCR with the optimized primers and PCR conditions. Amplified product was visualized on 10% PAGE stained with ethidium bromide, which resulted in the identification of 15 polymorphic markers. Of the 15 polymorphic markers, 10 markers were labeled with FAM fluorescent dye and genotyped in 3730 DNA Analyzer with HiDi formamide and GeneScan_TM- 500LIZ^®Size Standard (Applied Biosystems, U.S.). The results were analyzed with GeneMapper v4.0 software (Applied Biosystems, U.S.) to calculate the allele size and number of alleles. Genetic analysis was performed using Arlequin 3.1 [21], Genepop v4.0 [22] and Cervus v2.0 [11].

Three markers (Oosat55, Oosat59 and Oosat43) were selected based on the observed polymorphism between males and females. These markers were selected for inheritance study in the iso-female families of GMB4M. Pedigreed crosses were made to obtain F₁ and F₂ progeny while insects after mating, were preserved, for genotyping with these markers. At least two parental pairs were used to generate F₁ females that produced male and female progeny.