Lots of Accurately-aligned Proteins Initiated from Scratch
Often a critical first step in protein sequence analysis is obtaining a very large and accurate MSA. This is challenging considering that even relatively small alignments are often inaccurate. LAPIS uses a machine learning approach, in combination with other procedures, to align up to a million or more protein sequences with remarkable accuracy.
LAPIS executable (beta release):
A test set (of 10,000 GTPase sequences) as input to LAPIS:
Neuwald, A.F. LAPIS: Lots of accurately-aligned proteins initiated from scratch. In preparation.
National Institutes of Health, National Institute of General Medical Sciences grant R01GM125878.