2026-10-12T14:57:54Z https://www.scidb.cn/oai

10.57760/sciencedb.013tx 2026-09-28T16:58:02Z

2026-09-28 Data related to the DeepPlantTE article doi:10.57760/sciencedb.013tx en This dataset is associated with the manuscript “DeepPlantTE: a dedicated classification model for plant transposons” (manuscript ID: 20260299), submitted to Progress in Biochemistry and Biophysics. It contains plant transposon sequences from InpactorDB and MnTEdb, which were used as independent test datasets to evaluate the classification performance and generalization ability of DeepPlantTE. Neither dataset was used for model training. The InpactorDB test set contains 130,439 sequences from two classes, LTR/Copia and LTR/Gypsy. Because the range of classes represented in InpactorDB is relatively limited, we additionally used a subset of MnTEdb for supplementary testing. This subset comprises 4,077 sequences from five classes: LTR/Copia, LTR/Gypsy, DNA/TcMar, DNA/hAT, and LINE/L1. Testing on independent datasets from different sources and with different class compositions provides complementary evidence for assessing the model’s ability to classify plant transposon sequences and its generalization performance. independent test dataset; DeepPlantTE; plant transposon Jinshan Li PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset Science Data Bank