Physicial and chemical interactions, or contacts, between peptides are principally responsible for stabilizing a proteins structure and, consequently, determining its function. AlphaFold 2, developed by DeepMind, has proven capable of predicting protein structure with unprecedented accuracy. With predictions available for the majority of known proteins, AlphaFold has proven transformative for bioinformatics, resulting in an abundance of high-quality structural data. RING is a software that deterministically predicts non-covalent interactions in such structure files to create a residue interaction network (RIN). RINs created by RING capture high accuracy contact information with a wide verity of uses including: aiding in human understanding of protein structure or function, and training machine learning models. HomologyRing is a Python package that combines the functionality of RING with a homology search, enabling the analysis of contact conservation and variance across many of evolutionarily related proteins. Starting from a query structure or sequence, HomologyRing uses BLAST to perform a homology search against either UniProt or Protein Data Bank (PDB) databases, and subsequently using RING to collect contact information for corresponding AlphaFold or PDB structures. By mapping contacts to corresponding residues in a multiple sequence alignment (MSA), the pipeline at the core of HomologyRing synthesises a novel Homology enriched Residue Interaction Network (hRIN), and supplamentary tools included within the HomologyRing package aid in hRIN analysis. Using these tools, we demonstrate the utility of the resulting hRINS for characterizing how preservation and variance of contacts in homologs contributes to protein structure, function, and partner binding. HomologyRing compiles and visualizes detailed information on intra- and inter-chain contacts, and shows promise for a wide verity of potential applications, including: study of ligand, and partner binding specificity.

Physicial and chemical interactions, or contacts, between peptides are principally responsible for stabilizing a proteins structure and, consequently, determining its function. AlphaFold 2, developed by DeepMind, has proven capable of predicting protein structure with unprecedented accuracy. With predictions available for the majority of known proteins, AlphaFold has proven transformative for bioinformatics, resulting in an abundance of high-quality structural data. RING is a software that deterministically predicts non-covalent interactions in such structure files to create a residue interaction network (RIN). RINs created by RING capture high accuracy contact information with a wide verity of uses including: aiding in human understanding of protein structure or function, and training machine learning models. HomologyRing is a Python package that combines the functionality of RING with a homology search, enabling the analysis of contact conservation and variance across many of evolutionarily related proteins. Starting from a query structure or sequence, HomologyRing uses BLAST to perform a homology search against either UniProt or Protein Data Bank (PDB) databases, and subsequently using RING to collect contact information for corresponding AlphaFold or PDB structures. By mapping contacts to corresponding residues in a multiple sequence alignment (MSA), the pipeline at the core of HomologyRing synthesises a novel Homology enriched Residue Interaction Network (hRIN), and supplamentary tools included within the HomologyRing package aid in hRIN analysis. Using these tools, we demonstrate the utility of the resulting hRINS for characterizing how preservation and variance of contacts in homologs contributes to protein structure, function, and partner binding. HomologyRing compiles and visualizes detailed information on intra- and inter-chain contacts, and shows promise for a wide verity of potential applications, including: study of ligand, and partner binding specificity.

HomologyRing: A Python Tool for Analyzing Intra- and Inter-Chain Contact Specificity in Protein Families.

GRAVES, TANNER AARON
2023/2024

Abstract

Physicial and chemical interactions, or contacts, between peptides are principally responsible for stabilizing a proteins structure and, consequently, determining its function. AlphaFold 2, developed by DeepMind, has proven capable of predicting protein structure with unprecedented accuracy. With predictions available for the majority of known proteins, AlphaFold has proven transformative for bioinformatics, resulting in an abundance of high-quality structural data. RING is a software that deterministically predicts non-covalent interactions in such structure files to create a residue interaction network (RIN). RINs created by RING capture high accuracy contact information with a wide verity of uses including: aiding in human understanding of protein structure or function, and training machine learning models. HomologyRing is a Python package that combines the functionality of RING with a homology search, enabling the analysis of contact conservation and variance across many of evolutionarily related proteins. Starting from a query structure or sequence, HomologyRing uses BLAST to perform a homology search against either UniProt or Protein Data Bank (PDB) databases, and subsequently using RING to collect contact information for corresponding AlphaFold or PDB structures. By mapping contacts to corresponding residues in a multiple sequence alignment (MSA), the pipeline at the core of HomologyRing synthesises a novel Homology enriched Residue Interaction Network (hRIN), and supplamentary tools included within the HomologyRing package aid in hRIN analysis. Using these tools, we demonstrate the utility of the resulting hRINS for characterizing how preservation and variance of contacts in homologs contributes to protein structure, function, and partner binding. HomologyRing compiles and visualizes detailed information on intra- and inter-chain contacts, and shows promise for a wide verity of potential applications, including: study of ligand, and partner binding specificity.
2023
HomologyRing: A Python Tool for Analyzing Intra- and Inter-Chain Contact Specificity in Protein Families.
Physicial and chemical interactions, or contacts, between peptides are principally responsible for stabilizing a proteins structure and, consequently, determining its function. AlphaFold 2, developed by DeepMind, has proven capable of predicting protein structure with unprecedented accuracy. With predictions available for the majority of known proteins, AlphaFold has proven transformative for bioinformatics, resulting in an abundance of high-quality structural data. RING is a software that deterministically predicts non-covalent interactions in such structure files to create a residue interaction network (RIN). RINs created by RING capture high accuracy contact information with a wide verity of uses including: aiding in human understanding of protein structure or function, and training machine learning models. HomologyRing is a Python package that combines the functionality of RING with a homology search, enabling the analysis of contact conservation and variance across many of evolutionarily related proteins. Starting from a query structure or sequence, HomologyRing uses BLAST to perform a homology search against either UniProt or Protein Data Bank (PDB) databases, and subsequently using RING to collect contact information for corresponding AlphaFold or PDB structures. By mapping contacts to corresponding residues in a multiple sequence alignment (MSA), the pipeline at the core of HomologyRing synthesises a novel Homology enriched Residue Interaction Network (hRIN), and supplamentary tools included within the HomologyRing package aid in hRIN analysis. Using these tools, we demonstrate the utility of the resulting hRINS for characterizing how preservation and variance of contacts in homologs contributes to protein structure, function, and partner binding. HomologyRing compiles and visualizes detailed information on intra- and inter-chain contacts, and shows promise for a wide verity of potential applications, including: study of ligand, and partner binding specificity.
Proteins
Contacts
Homology
Python
AlphaFold
File in questo prodotto:
File Dimensione Formato  
Graves-pdf1b.pdf

accesso aperto

Dimensione 5.44 MB
Formato Adobe PDF
5.44 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/80891