We propose a feature vector approach to characterize the variation in large data sets of biological sequences. Each candidate sequence produces a single feature vector constructed with the number and
Amino acid sequence of the VL domains were acquired from the Amyloid Light Chain Database (ALBase Boston university) using the patient ID’s as search. Amino acid differences between LEN and SMA (total
Supporting data for reproducing figures and using toolkit in the manuscript ICEPIC: A Toolkit to Discover Ice Binding Proteins from SequencePlease see https://github.com/netrias/ICEPIC for more inform