On the Salient Limitations of `On the Salient Limitations of the Methods of Assembly Theory and their Classification of Molecular Biosignatures' (2403.12452v3)

Abstract: Assembly Theory (AT) is a theory that explains how to determine if a complex object is the product of evolution. Here we explain why attempts to compare AT to compression algorithms, ref 1, does not help identify if the object is the product of selection or not. Specifically, we show why aims to perform benchmark comparisons of different compression schemes to compare the performance of Molecular Assembly Indices against standard compression schemes in determining living vs. non-living samples fails to classify the data correctly. In their approach, Uthamacumaran et al., ref 1, compress data from Marshall et al.2 describing the experimental basis for assembly theory and evaluate the difference in the resulting distributions of data. After several computational experiments Uthamacumaran et al., ref 1, conclude that other compression techniques obtain better results in the problem of detecting life and non-life that of assembly pathways. Here we explain why the approach presented by Uthamacumaran et al.1 appears to be lacking, and why it is not to reproduce their analysis with the information given.