Update README.md

README.md

@@ -4,4 +4,9 @@ task_categories:
 - feature-extraction
 size_categories:
 - 100M<n<1B
----
+---
+
+This dataset is collected from [TCRdb2.0](https://guolab.wchscu.cn/TCRdb2/). 
+The original sequence records are 691,744,135 with redundancy. 
+`tcrdb2.txt` removes duplicates and remains 290,314,598.
+`tcrdb2CDhit75.txt` removes sequences with less than 75 identity and remains 1,782,927.