Related Experiment Video
Updated: Jun 13, 2026
08:57
Using Phylogenetic Analysis to Investigate Eukaryotic Gene Origin
Published on: August 14, 2018
Phylogenetic corrections and higher-order sequence statistics in protein families: Potts vs multiple sequence
Kisan Khatri1, Ronald M Levy2, Allan Haldane1
1Department of Physics and Center for Biophysics and Computational Biology, Temple University, Philadelphia, Pennsylvania 19122, USA.
Physical Review Research
|June 12, 2026
Summary
Physics-based Potts models outperform machine learning models like MSA Transformer in identifying protein biophysical constraints when accounting for evolutionary relationships. Both models generate sequences predicted to fold into native-like structures.
Area of Science:
- Computational biology
- Protein sequence analysis
- Machine learning in bioinformatics
Background:
- Generative machine learning models, including Potts models and MSA Transformer (MSA-T), are used for protein multiple sequence alignment (MSA).
- These models aim to capture biophysical constraints within proteins by reproducing MSA statistics.
- The ability of models to capture complex interactions beyond pairwise residue-residue terms and account for phylogenetic structure is under investigation.
Purpose of the Study:
- To compare the performance of the Potts model and MSA-T in reconstructing higher-order sequence statistics.
- To investigate the impact of phylogenetic relationships on model performance in detecting biophysical sequence constraints.
- To evaluate the ability of generated protein sequences to fold into nativelike structures.
Main Methods:
- Comparison of Potts model and MSA Transformer (MSA-T) performance on protein multiple sequence alignment (MSA) datasets.
- Analysis of higher-order sequence statistics and biophysical sequence constraints.
- Implementation of explicit corrections for phylogenetic dependencies in MSAs.
- Structure prediction of generated sequences using AlphaFold.
Main Results:
- Model performance is highly dependent on the treatment of phylogenetic relationships within MSAs.
- The Potts model, with explicit phylogenetic corrections, outperforms MSA-T in detecting biophysical epistatic interactions.
- Sequences generated by both models are predicted to fold into nativelike structures by AlphaFold.
Conclusions:
- Explicitly addressing phylogenetic dependencies is crucial for accurately modeling biophysical constraints in proteins.
- Potts models demonstrate superior capability in identifying complex, evolutionarily-informed biophysical interactions compared to MSA-T.
- Both modeling approaches show promise in generating functional protein sequences.
Related Concept Videos
02:54Evolutionary Relationships through Genome Comparisons
Genome comparison is one of the excellent ways to interpret the evolutionary relationships between organisms. The basic principle of genome comparison is that if two species share a common feature, it is likely encoded by the DNA sequence conserved between both species. The advent of genome sequencing technologies in the late 20th century enabled scientists to understand the concept of conservation of domains between species and helped them to deduce evolutionary relationships across diverse...
02:47Protein Families
Protein families are groups of homologous proteins; that is, they have similarities in amino acid sequences and three-dimensional structures. Protein families usually occur because of gene duplication, where an additional copy of a gene is inserted into the genome of an organism. Mutations that change the amino acids but still allow the protein to be properly synthesized, will lead to new protein family members. If these new proteins contain similar amino acids in key locations, protein...
02:47Protein Families
Protein families are groups of homologous proteins; that is, they have similarities in amino acid sequences and three-dimensional structures. Protein families usually occur because of gene duplication, where an additional copy of a gene is inserted into the genome of an organism. Mutations that change the amino acids but still allow the protein to be properly synthesized, will lead to new protein family members. If these new proteins contain similar amino acids in key locations, protein...
01:28Microbial Phylogeny
Understanding the evolutionary relationships among microorganisms is fundamental to microbial ecology and taxonomy. Phylogenetic trees are essential tools for inferring these relationships, relying primarily on comparative analyses of molecular sequences such as DNA, RNA, or proteins. In microbial studies, these trees typically depict the evolutionary paths of diverse bacterial and archaeal species by mapping genetic differences accumulated over time.Phylogenetic trees are composed of tips,...
03:21Phylogenetic Trees
Phylogenetic trees come in many forms. It matters in which sequence the organisms are arranged from the bottom to the top of the tree, but the branches can rotate at their nodes without altering the information. The lines connecting individual nodes can be straight, angled, or even curved.The length of the branches can depict time or the relative amount of change among organisms. For instance, the branch length might indicate the number of amino acid changes in the sequence that underlies the...
03:21Phylogenetic Trees
Phylogenetic trees come in many forms. It matters in which sequence the organisms are arranged from the bottom to the top of the tree, but the branches can rotate at their nodes without altering the information. The lines connecting individual nodes can be straight, angled, or even curved.The length of the branches can depict time or the relative amount of change among organisms. For instance, the branch length might indicate the number of amino acid changes in the sequence that underlies the...