Sequences with missing amino acids/incorrect ANARCI extraction contain the string "nan"

[MiguelHK]

Hello! I would like kindly inform that I have found some sequences in AntiRef that contain the string "nan". I believe this is due to how the sequences are reconstructed with pandas. The total number of sequences with the string "nan" are:

• AntiRef100: 21,284
• AntiRef90: 10,335

The code used to find these sequences was the following:
grep 'nan' /home/ubuntu/MHK/antiref/<antiref_file> | grep -v '^>' | wc -l