Abstract: Deep bio-information analysis was carried out using transcriptome data in order to develop SSR marks in hybrid hazelnut Corylus heterophylla × C. avellana. In total，35 604 SSR sequences were found in 25 649 SSR containing Unigenes. In all the SSR sequences，mono-nucleotide，dinucleotide and trinucleotide repeats belonged to high frequent type，and their number and percent were 10 102（28.37%），16 543（46.46%）and 7 035（19.76%）respectively. In mono-nucleotide repeats，A/T was most abundant，accounting for 28.13% of the total SSR amount. In dinucleotide repeats，AG/CT，AT/AT，AC/GT were more abundant，accounting for 35.79%，7.47% and 3.17% of the total. In repeats of trinucleotide，tetranucleotide，pentanucleotide and hexanucleotide，the highest percent sequence of each nucleotide repeat were AAG/CTT，AAAG/CTTT，AAAAG/CTTTT，AAAAAG/CTTTTT，and they accounted for 7.24%，0.33%，0.38% and 0.15% of the total SSR amount respectively. Sixty pairs of randomly chosen SSR primers were synthesized and amplified from 25 Corylus samples，including Corylus heterophylla Fisch，C. heterophylla × C. avellana，C. avellana and C. mandshurica Maxim. Polymorphism information content（PIC）values of seven polymorphic pairs of primers varied from 0.35 to 0.70，and 52 polymorphic bands were obtained. UPGMA（Unweighted Pair Group Method Analysis）clustering results indicated that these primers could precisely separate germplasms mentioned above into different groups. This study clarifies SSR distribution characteristics of hybrid hazelnut C. heterophylla × C. avellana and provides a scientific basis for further studies of genetic linkage map construction in these species.

Key words: hazelnut, transcriptome, simple sequence repeat, germplasm identification, primer