Article · 1 Apr 2024
Conference abstract
A shortcut from large-scale genomic data to protein structure and function
Abstract
Select any words in the title or abstract, or press this and choose a sentence.
Statement of the Problem: With 90 million proteins in the public databases and an unprecedented growth, only robust unsupervised and automated methods can realistically achieve the comprehensive functional annotations of the protein space. Importantly, in recent years, the accumulation of new sequences is mostly driven by deep sequencing efforts and a high-quality assembly phase. To address the pressing challenge of accurate annotation, we offer a classification resource that is created by an unsupervised analysis of full-length protein sequences. The charting of the protein space to families is model free with a quality that matches the state-of-the-art expert systems. Methodology & Theoretical Basis: Machine learning (ML) methods are becoming fundamental in annotating unknown proteins and their heterogeneous properties. The quality of automatic charting the protein space is assessed by the structural conservation measures which provide an unavoidable bridge to function. The entire protein space is created using hierarchical clustering. We address the basis for detecting overlooked connections between families; the division of families to subfamilies and the requirement for a reliable inference. Findings: The ProtoNet platform that includes analytical tools to explore the protein space will be presented included its version for genome scale analyses. An innovative structural biology approach focuses on determining residue-level protein properties, such as sites of post-translational modifications (PTMs). We formulated it in ML method for a residue-level property called ASAP. It extracts numerous features from raw sequences and supports easy integration of external features such as secondary structure, solvent accessibility, intrinsically disorder or PSSM profiles. Features are then used to train ML classifiers. ASAP can create new classifiers within minutes for a variety of tasks, including PTM prediction. Conclusion & Significance: We present a set of methodologies for bridging protein function at different levels from single residue to a full proteins and genome. These tools and platforms match state-of-the-art results and shed light on unexplored evolutionary processed that cross all domains of life. We conclude that despite minimal sequence similarity fingerprints of evolutionary processes remain powerful for analyzing new proteomes at a genomic scale, for protein design tasks, mass spectrometry search engines and the discovery of new bioactive proteins. PDF HTML
Keywords
Ask about this article
Answers come only from this article's title, abstract and keywords, each sentence pointing to where it comes from. Your question is sent to OpenAI to write the answer. Scholars Central does not store it.
Similar by topic
Perspective · 2 Jun 2022
Integrating Genomics Improves Large-Scale of Human Genomics
Research article · 15 Aug 2016
Genome-wide Screening of Cell Wall-related Modules in Mycobacterium tuberculosis based on Large-scale Co-Expression Analysis
Research article · 2014
Large-scale data classification based on clustering feature tree decomposition
Research article · Nov 2012
Speeding Up Large-Scale Next Generation Sequencing Data Analysis with pBWA
More from this journal
The newest in Biochemistry & Analytical Biochemistry.
Article · 26 Dec 2025
Devising a Treatment Plan considering the Impact of Thermoregulated Feature of Salvia sclarea (Clary Sage) Essential Oil. The Cases of Hot-flashes with Hypertension and Menopause.
Article · 26 Dec 2025
Mathematical Descriptions of Enzyme–Substrate Interactions
Article · 26 Dec 2025
Dynamic Principles of Enzyme Reaction Rates in Biological Systems
Conferences in this field
Upcoming conferences in this field.
Follow
New articles from what you follow go into your email digest, once you agree to alert emails.
- JournalBiochemistry & Analytical Biochemistry
- SubjectBiochemistry & Biotechnology
- SubjectChemistry
- BylineArticles whose byline prints “Michal Linial”
