Ten rice genotypes were evaluated for both morphological and molecular studies. The field experiment was conducted in perforated small circular plastic pots (16 cm in length and 12 cm in diameter). The seeds were sown in seedbed on 28 January and then 30-day-old seedlings were transplanted into the perforated pots. At first, three seedlings were transplanted in each pot and later thinned into one. Net clothes were used for preparation of bags, which were well fitted in pots. Inside the pot, a net bag was placed. Then the pots were filled with soil containing N-, P-, and M-containing fertilizers. The soil was fertilized with 50 mg N, 25 mg P, and 25 mg K per kilogram (kg) of soil. Trays (130 cm × 75 cm × 25  cm) were filled with normal tap water and pots were placed into the trays. These trays served as water bath. Four trays were set for this study; each one contained 50 pots.

Four treatments with different durations of submergence were used for rice genotypes to observe variable responses and the performance of yield and yield-contributing characters. The genotypes were submerged in the vegetative stage for 7 and 14 days, and only for 7 days in the reproductive stage. A set of genotypes was maintained in natural growth conditions without undergoing the submergence treatment termed as control. Artificial submerged condition was created in a large tank of 100 cm with water. Tanks were filled with normal tap water. When plants rose to the vegetative (8–10 tiller) and reproductive stages, the water from the trays was removed and the pots were placed into large tanks. Fifty plants were evaluated for submergence for seven days in the vegetative stage, 50 plants were evaluated for submergence for 14 days in the vegetative stage and 50 plants were evaluated for submergence at the reproductive stage for seven days. For each treatment, five replications were made.

The floodwater conditions in the tank were monitored every morning and evening. The temperature of water was measured by a thermometer and the pH was measured with a pH meter. Dissolved oxygen was measured every morning and evening with a dissolved oxygen meter (DO 5509, Lutron Enterprise, Taiwan).

The Standard Evaluation System (SES) developed by the International Rice Research Institute (IRRI) [30] (Table 2) was used for scoring the rice genotypes.

To know the performance of the rice genotypes at different levels of submergence and growth stages, data were recorded for days to flowering, plant height, tiller number plant⁻¹, effective tiller plant⁻¹, panicle length, dry matter, harvest index, yield plant⁻¹, chlorophyll content and so on. The days to flowering were recorded as days from sowing to flowering when 50% of the plants of each tray flowered. The plant height was measured in centimetres from main culms from the ground level to the tip of its panicle. The total number of tillers (effective and non-effective) were counted for each sample plant. Panicle length was recorded as the distance from the last node of the rachis to tip of the main panicle, which was randomly selected for each plant; the average was taken.

The dry weight (g) of each sample was taken. The samples were oven dried at 50 °C for three days. From every sample plant, 100 clean and sun-dried seeds were counted, and the average was taken and adjusted at 14% seed moisture content. The harvest index was recorded for each sample by dividing the economic yield by the biological yield. The total grain weight (g) of each sample plant was taken after cleaning and sun-drying the samples; the average was taken and adjusted at 14% seed moisture content. The chlorophyll content was measured using chlorophyll meter (SPAD-502 Plus, Konica Minolta, Europe). It is a non-destructive method by which chlorophyll can be measured from the healthy leaves of existing plants. Three leaves of each plant were selected and the average value was taken. The chlorophyll content was measured both before and after submergence.

The morphological data was analysed by the pieces of software MSTATC and MINITAB 17.

Ten rice genotypes were analysed with three Single Sequence Repeat (SSR) markers (Table 3); among them two were submergence gene specific markers and one was a diversity marker.

Genomic DNA was isolated following the modified CTAB method [31]. Plants were grown in small trays (33 cm × 29 cm × 12 cm) for extracting DNA from leaves. Twenty-one-day-old leaves were collected from growing rice plants and used for DNA extraction.

PCR was performed for DNA amplification. For each sample, Taq buffer (B) 1.00 μl, MgCl₂ 0.60 μl, dNTPs 1.00 μl, Taq polymerase 0.12 μl, primer forward 1.00 μl, primer reverse 1.00 μl, ddH₂O 3.28 μl were used. Two μl sample DNA was used for each reaction. Amplification with PCR was performed using the following conditions: denaturation at 94 °C for 5 min; 35 cycles of 1 min denaturation at 94 °C, 1 min annealing at 55 °C (for SC3 and RM24) and 57 °C (for SUB1), 2 min extension at 72 °C and a final extension at 72 °C for 7 min. The SSR amplification products were separated in a vertical denaturing 8% polyacrylamide gel. DNA fragments were revealed using ethidium bromide staining. The gels were stained for 20–25 min and were documented with a gel documentation system.

The size of the amplified fragments was determined by comparing the migration distance of amplified fragments relative to the molecular weight of the known-size marker commonly known as ladder, 100 base pair (bp) and 20 bp ladder using Alpha-Ease FC 5.0 software (Alpha Innotech, USA). The number of allele per locus, major allele frequency, gene diversity, PIC value and Nei's [32] genetic distance and UPGMA dendrogram were calculated and constructed using Power Marker version 3.25 [33]. All the genotypes were scored for the presence and absence of the SSR bands throughout all 10 genotypes, and the data were exported to binary data for presence (1) and absence (0) or as missing observation for further analysis with NTSYS-pc version 2.2. NTSYS-pc was used to construct UPGMA dendrogram showing the distance based interrelationship among the rice genotypes.

A decline in the performance of rice genotypes was observed in submerged conditions. Most of the genotypes, either tolerant check or SUB1introgressed advanced lines or susceptible varieties, showed similar performance in control conditions. Days to flowering increased due to giving submergence treatment to the genotypes at a 7-day vegetative stage, whereas days to flowering decreased at 14 days vegetative and 7 days reproductive stages; the mean ranged from 36 to 120 days (Table 5). Genotypes showed a minimum mean number of days to flowering at 7 days in submerged conditions in the reproductive stage and a maximum mean number in control conditions. The yield-contributing characters also exhibited the declining pattern after submerged treatment. In control conditions, no significant difference was observed among the genotypes. However, the tolerant check and the SUB1 gene introgressed lines showed considerably better performance in submerged conditions compared to the other genotypes. A significant genotype × environment interaction was evidenced in this study as the environment had a great effect on genotypes. The plant height of experimental rice genotypes ranged from 29 to 118 cm; in control conditions the results did not show remarkable differences among the genotypes, whereas in submerged conditions the plant height was reduced among the genotypes remarkably. The plant height decreased mostly at 14 days in the vegetative stage and at 7 days in the reproductive stage, with reduced average heights of 64 and 40 cm, respectively. Binadhan-11, Binadhan-12, RC 192, RC 193, RC 227, RC 249 and RC 251 showed better growth performance among the genotypes in submerged conditions. A decrease in the number of tillers was also observed in this study due to submergence (Table 5).

The numbers of tillers plant⁻¹ were reduced at about 4, 10 and 8 for 7 and 14 days in the vegetative stage and 7 days in the reproductive stage under submerged conditions, respectively. Also, the tolerant check and SUB1 introgressed line performed better compared to the susceptible check. Effective tillers plant⁻¹ were also declined after the submergence treatment and ranged from 1 to 14. Binadhan-11, Binadhan-12, RC 249 and RC 251 showed comparatively better performance in both control and submerged conditions (Table 5). The panicle length was reduced to about 2 cm, 11 cm and 7 cm for 7 and 14 days in the vegetative stage and 7 days in the reproductive stage respectively under submerged conditions (Table 6). In control conditions, the tolerant check, SUB1 introgressed lines and susceptible genotypes exhibited no distinct reduction of the panicle length. The reduction was more drastic at 14 days in the vegetative stage under submerged conditions. Above ground dry weight also reduced due to submergence treatment and the mean value ranged from 2.04 g to 65.69 g (Table 6). The plants showed a maximum reduction in their dry weight at 14 days in the vegetative stage under submerged conditions.

Binadhan-11, Binadhan-12, RC 249 and RC 251 showed better performance in submerged condition.

The weights of 100 seeds were also taken for each of the treated genotypes and the result showed a little variation among the treatments. However, the seed weight decreased due to submergence at 14 days in the vegetative stage and 7 days in the reproductive stage; the reduction was about 1 g (Table 6). The yield plant⁻¹ was also reduced drastically under submerged conditions. The variation in yield in control condition among the genotypes was genetically inherited. All the genotypes showed a reduction in yield in submerged conditions. The reduction was 15 g, 31 g and 23 g at 7 and 14 days in the vegetative stage and 7 days in the reproductive stage, under submerged conditions, respectively (Table 6). The reduction was more in susceptible genotypes compared to the tolerant check SUB1 introgressed lines. In 14 days of submergence in the vegetative stage, genotypes gave minimum yield, but Binadhan-11, Binadhan-12, RC 249 and RC 251 showed better yield performance than other genotypes (Figs. 1 and 2).

The harvest index was calculated from grain yield and dry weight in the present study. However, the harvest index was similar for maximum genotypes in control condition, whereas it declined in submerged conditions. The decrease was maximum at 14 days in the vegetative stage under submerged conditions.

The chlorophyll content was measured both for control and submerged conditions. There were no visible differences in control conditions among the genotypes. The chlorophyll content was reduced a little at 7 days in submerged condition in the vegetative stage, though in some cases it increased. The chlorophyll content was reduced by 19 and 15 SPAD (Soil-Plant Analyses Development) units, respectively, at 14 days in the vegetative stage and 7 days in the reproductive stage, under submerged conditions (Table 7).

A wide range of variations was observed among the rice genotypes for the studied traits. The phenotypic coefficient of variation (PCV) was higher than the corresponding genotypic coefficient of variation (GCV) for all traits (Table 8), which also indicates that all are interrelated to some extent. The values showed that the yields plant⁻¹ (54.92 and 57.86) have the highest GCV and PCV. The dry weight showed GCV and PCV (52.01 and 54.81) close to the yield plant⁻¹. Effective tiller plant⁻¹ (34.83 and 37.22) showed GCV and PCV close to the tiller number plant⁻¹ (32.74 and 36.45), plant height (31.02 and 35.74) and hundred seed weight (33.06 and 34.25). The chlorophyll content (23.55 and 25.03), the number of days to flowering (29.05 and 30.38), the panicle length (28.82 and 30.62) also showed very close GCV and PCV (Table 8). These results indicate that high values of the genotypic coefficient of variation and of the phenotypic coefficient of variation have the possibility to improve the yield through the selection of that trait.

The estimates of heritability act as a predictive instrument in expressing the reliability of the phenotypic value. The high heritability is very effective for the selection of a particular trait. Among the traits, the highest heritability was observed for 100 seed weights (93.16) followed by days to flowering (91.43), yield plant⁻¹ (90.08), dry weight (90.04), panicle length (88.64), chlorophyll content (88.54), effective tiller (87.58), tiller⁻¹ (80. 68) and plant height (75.35) (Table 8). The genetic advance is a useful indicator of the progress that can be expected as a result of exercising selection on the pertinent population. Good genetic advance along with high heritability was found for days to flowering (45.55) and plant height (40.05) (Table 8).

Using three SSR marker-based primers were used across 10 genotypes and 19 alleles were identified among the studied genotypes. The loci SC3 had the highest number of alleles (8), whereas the loci SUB1 had the lowest number of alleles (5). The average value of the allele was 6.333 (Table 9). The highest genetic diversity was found for SC3 (0.840) and the lowest for SUB1 (0.720); see Table 9. In molecular analysis, microsatellite loci are also multi allelic and the alleles are co-dominant, suggesting their relative superiority in detecting DNA polymorphism. On an average, 33.3% of the 10 genotypes shared a common major allele ranging from 30% (SC3 and RM 24) to 40% (SUB1) at each locus (Table 9). The polymorphic information contents the PIC value, which reflects the allelic diversity and the frequency among the varieties. The PIC value of each marker can be evaluated based on of its alleles, and the PIC values varied significantly for all the studied SSR loci. In the present study, the level of polymorphism among the 10 genotypes was evaluated by calculating the PIC values for each one of the three SSR loci. The PIC values ranged from 0.676 (SUB1) to 0.823 (SC3), with an average of 0.757 per locus (Table 9).

Genomic banding patterns of 10 rice genotypes using three SSR markers are shown in Figs. 3 to 5. The primer RM24, which is a diversity primer, showed polymorphism among the genotypes used in this experiment. Along with two submergence specific primers, SUB1 and SC3 were also evaluated in this study for the comparison of the submergence tolerance of 10 rice genotypes. The genotypes, which showed genomic bands with the same position or near with the submergence tolerant variety Binadhan-11 and Binadhan-12, were supposed to be tolerant to submergence.

The Unweighted Pair Group Method with Arithmetic means (UPGMA) method was used for cluster analysis to differentiate the studied genotypes into groups based on similarity coefficients. Three clusters were made for the 10 rice genotypes based on their different levels of submergence. However, cluster 1 was divided into two sub-clusters. The first sub-cluster is comprised of BRRI dhan28, RC 225 and Binadhan-12, whereas sub-cluster 2 consists of RC 227 and RC 249. The second cluster was also divided into two sub-clusters: the first one consists of RC 193, RC 251, RC 192 and the second is comprised of Binadhan-11. The third cluster is made up with only one rice genotype, Binadhan-7. The genotypes that showed comparatively the same banding pattern were grouped into the same cluster (Fig. 6).