Anastasia Sveshnikova
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Anastasia Sveshnikova has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Vassily Hatzimanikatis
Swiss Federal Institute of Technology (EPFL)
6 shared publications - Homa MohammadiPeyhani
Prescient Design, Genentech
3 shared publications - Omid Oftadeh
Laboratory of Computational Systems Biotechnology, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
2 shared publications - Alan Bridge
SIB Swiss Institute of Bioinformatics, Centre Medical Universitaire, 1 rue Michel Servet, CH-1211, Geneva, Switzerland
1 shared publication - Anne Morgat
Swiss-Prot Group, SIB Swiss Institute of Bioinformatics, CMU, 1 rue Michel-Servet, CH-1211 Geneva 4, Switzerland
1 shared publication - Catherine Rivoire
Swiss-Prot Group, SIB Swiss Institute of Bioinformatics, CMU, 1 rue Michel-Servet, CH-1211 Geneva 4, Switzerland
1 shared publication
Latest Publications
- Open Access
- 24 May 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-13
Designing pathways for bioproducing complex chemicals by combining tools for pathway extraction and ranking
The synthesis of many important biochemicals involves complex molecules and numerous reactions. The design and optimization of whole-cell biocatalysts for the production of these molecules requires metabolic modeling to extract production pathways from biochemical databases and integrate them into genome-scale models of the host. However, the synthesis of such complex molecules often requires reactions from multiple pathways operating in balanced subnetworks that are not assembled in existing databases. Here, we present SubNetX, a computational algorithm that extracts reactions from a database and assembles balanced subnetworks to produce a target biochemical from selected precursor metabolites, energy currencies, and cofactors. These subnetworks can be integrated into whole-cell models, allowing the reconstruction and ranking of alternative biosynthetic pathways based on yield, length, and other design goals. We apply SubNetX to 70 industrially relevant natural and synthetic chemicals to demonstrate the application of this pipeline.
- 24 May 2025
- Published by Springer Nature in Nature Communications
- Vol. 16 (1) , 1-13
Open AccessDesigning pathways for bioproducing complex chemicals by combining tools for pathway extraction and ranking
The synthesis of many important biochemicals involves complex molecules and numerous reactions. The design and optimization of whole-cell biocatalysts for the production of these molecules requires metabolic modeling to extract production pathways from biochemical databases and integrate them into genome-scale models of the host. However, the synthesis of such complex molecules often requires reactions from multiple pathways operating in balanced subnetworks that are not assembled in existing databases. Here, we present SubNetX, a computational algorithm that extracts reactions from a database and assembles balanced subnetworks to produce a target biochemical from selected precursor metabolites, energy currencies, and cofactors. These subnetworks can be integrated into whole-cell models, allowing the reconstruction and ranking of alternative biosynthetic pathways based on yield, length, and other design goals. We apply SubNetX to 70 industrially relevant natural and synthetic chemicals to demonstrate the application of this pipeline.
- Open Access
- 9 September 2024
- Published by Springer Nature in Scientific Data
- Vol. 11 (1) , 1-19
EnzChemRED, a rich enzyme chemistry relation extraction dataset
Expert curation is essential to capture knowledge of enzyme functions from the scientific literature in FAIR open knowledgebases but cannot keep pace with the rate of new discoveries and new publications. In this work we present EnzChemRED, for Enzyme Chemistry Relation Extraction Dataset, a new training and benchmarking dataset to support the development of Natural Language Processing (NLP) methods such as (large) language models that can assist enzyme curation. EnzChemRED consists of 1,210 expert curated PubMed abstracts where enzymes and the chemical reactions they catalyze are annotated using identifiers from the protein knowledgebase UniProtKB and the chemical ontology ChEBI. We show that fine-tuning language models with EnzChemRED significantly boosts their ability to identify proteins and chemicals in text (86.30% F1 score) and to extract the chemical conversions (86.66% F1 score) and the enzymes that catalyze those conversions (83.79% F1 score). We apply our methods to abstracts at PubMed scale to create a draft map of enzyme functions in literature to guide curation efforts in UniProtKB and the reaction knowledgebase Rhea.
- 9 September 2024
- Published by Springer Nature in Scientific Data
- Vol. 11 (1) , 1-19
Open AccessEnzChemRED, a rich enzyme chemistry relation extraction dataset
Expert curation is essential to capture knowledge of enzyme functions from the scientific literature in FAIR open knowledgebases but cannot keep pace with the rate of new discoveries and new publications. In this work we present EnzChemRED, for Enzyme Chemistry Relation Extraction Dataset, a new training and benchmarking dataset to support the development of Natural Language Processing (NLP) methods such as (large) language models that can assist enzyme curation. EnzChemRED consists of 1,210 expert curated PubMed abstracts where enzymes and the chemical reactions they catalyze are annotated using identifiers from the protein knowledgebase UniProtKB and the chemical ontology ChEBI. We show that fine-tuning language models with EnzChemRED significantly boosts their ability to identify proteins and chemicals in text (86.30% F1 score) and to extract the chemical conversions (86.66% F1 score) and the enzymes that catalyze those conversions (83.79% F1 score). We apply our methods to abstracts at PubMed scale to create a draft map of enzyme functions in literature to guide curation efforts in UniProtKB and the reaction knowledgebase Rhea.
- Open Access
Designing pathways for bioproducing complex chemicals by combining tools for pathway extraction and ranking
The synthesis of many important biochemicals involves complex molecules and many reactions. Therefore, the design and optimization of whole-cell biocatalysts to produce these molecules requires the use of metabolic modeling. Such modeling involves the extraction of the production pathways from biochemical databases and their integration into genome-scale metabolic models of the host organism. However, the synthesis of such complex molecules requires reactions from multiple pathways operating in balanced subnetworks that are not assembled in existing databases. Here we present SubNetX, a novel computational algorithm that extracts reactions from a given reaction database and assembles balanced reaction subnetworks to produce a target biochemical from a selected set of precursor metabolites, energy currencies, and cofactors of the host organism. These subnetworks can be directly integrated into whole-cell metabolic models, and using available methods, we can then reconstruct all alternative biosynthetic pathways and rank them according to design criteria such as yield, pathway length, and other optimization goals. We applied SubNetX to eight selected secondary metabolites and one non-natural chemical used as pharmaceuticals to demonstrate the potential of this pipeline.
Open AccessDesigning pathways for bioproducing complex chemicals by combining tools for pathway extraction and ranking
The synthesis of many important biochemicals involves complex molecules and many reactions. Therefore, the design and optimization of whole-cell biocatalysts to produce these molecules requires the use of metabolic modeling. Such modeling involves the extraction of the production pathways from biochemical databases and their integration into genome-scale metabolic models of the host organism. However, the synthesis of such complex molecules requires reactions from multiple pathways operating in balanced subnetworks that are not assembled in existing databases. Here we present SubNetX, a novel computational algorithm that extracts reactions from a given reaction database and assembles balanced reaction subnetworks to produce a target biochemical from a selected set of precursor metabolites, energy currencies, and cofactors of the host organism. These subnetworks can be directly integrated into whole-cell metabolic models, and using available methods, we can then reconstruct all alternative biosynthetic pathways and rank them according to design criteria such as yield, pathway length, and other optimization goals. We applied SubNetX to eight selected secondary metabolites and one non-natural chemical used as pharmaceuticals to demonstrate the potential of this pipeline.
- Open Access
Enzyme promiscuous profiles for protein sequence and reaction annotation
1.1 Novel sequencing techniques and biochemical pathway prediction resources provide a wealth of data on novel proteins and computationally predicted enzymatic reactions. Accurate matching of protein sequences to enzymatic activities is crucial for advancing synthetic biology and metabolic engineering efforts. Here we present BridgIT+, a computational workflow that accounts for enzyme promiscuity and accurately predicts protein-reaction and reaction-protein associations. BridgIT+ builds upon the promiscuity-based method for annotating orphan and novel reactions with enzymatic activities, BridgIT, and utilizes position-specific scoring matrices (PSSM). The framework uses sequence alignment and enzyme promiscuity predictions to analyze protein sequences, identify sequence patterns, and create promiscuous protein sequence profiles for each reaction. These profiles allow us to predict the protein sequences most likely involved in the reaction. We showcase BridgIT+ by annotating (i) computationally predicted reactions with proteins and (ii) unannotated proteins of E. coli proteome with enzymatic functions. We demonstrated the performance of BridgIT+ on several biochemical assays and compared it to three current state-of-the-art methods for matching proteins and reactions. We anticipate that the proposed conceptual framework will enhance our understanding of gene-protein-reaction relations and advance biological sequence and reaction annotation in biology and synthetic biology studies.
Open AccessEnzyme promiscuous profiles for protein sequence and reaction annotation
1.1 Novel sequencing techniques and biochemical pathway prediction resources provide a wealth of data on novel proteins and computationally predicted enzymatic reactions. Accurate matching of protein sequences to enzymatic activities is crucial for advancing synthetic biology and metabolic engineering efforts. Here we present BridgIT+, a computational workflow that accounts for enzyme promiscuity and accurately predicts protein-reaction and reaction-protein associations. BridgIT+ builds upon the promiscuity-based method for annotating orphan and novel reactions with enzymatic activities, BridgIT, and utilizes position-specific scoring matrices (PSSM). The framework uses sequence alignment and enzyme promiscuity predictions to analyze protein sequences, identify sequence patterns, and create promiscuous protein sequence profiles for each reaction. These profiles allow us to predict the protein sequences most likely involved in the reaction. We showcase BridgIT+ by annotating (i) computationally predicted reactions with proteins and (ii) unannotated proteins of E. coli proteome with enzymatic functions. We demonstrated the performance of BridgIT+ on several biochemical assays and compared it to three current state-of-the-art methods for matching proteins and reactions. We anticipate that the proposed conceptual framework will enhance our understanding of gene-protein-reaction relations and advance biological sequence and reaction annotation in biology and synthetic biology studies.
- Open Access
- 4 April 2022
- Published by Elsevier in Metabolic Engineering
- Vol. 72, 259-274
ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds
- 4 April 2022
- Published by Elsevier in Metabolic Engineering
- Vol. 72, 259-274
Open AccessARBRE: Computational resource to predict pathways towards industrially important aromatic compounds
- Open Access
- 23 March 2022
- Published by Springer Nature in Nature Communications
- Vol. 13 (1) , 1-12
Expanding biochemical knowledge and illuminating metabolic dark matter with ATLASx
Metabolic “dark matter” describes currently unknown metabolic processes, which form a blind spot in our general understanding of metabolism and slow down the development of biosynthetic cell factories and naturally derived pharmaceuticals. Mapping the dark matter of metabolism remains an open challenge that can be addressed globally and systematically by existing computational solutions. In this work, we use 489 generalized enzymatic reaction rules to map both known and unknown metabolic processes around a biochemical database of 1.5 million biological compounds. We predict over 5 million reactions and integrate nearly 2 million naturally and synthetically-derived compounds into the global network of biochemical knowledge, named ATLASx. ATLASx is available to researchers as a powerful online platform that supports the prediction and analysis of biochemical pathways and evaluates the biochemical vicinity of molecule classes (https://lcsb-databases.epfl.ch/Atlas2).
- 23 March 2022
- Published by Springer Nature in Nature Communications
- Vol. 13 (1) , 1-12
Open AccessExpanding biochemical knowledge and illuminating metabolic dark matter with ATLASx
Metabolic “dark matter” describes currently unknown metabolic processes, which form a blind spot in our general understanding of metabolism and slow down the development of biosynthetic cell factories and naturally derived pharmaceuticals. Mapping the dark matter of metabolism remains an open challenge that can be addressed globally and systematically by existing computational solutions. In this work, we use 489 generalized enzymatic reaction rules to map both known and unknown metabolic processes around a biochemical database of 1.5 million biological compounds. We predict over 5 million reactions and integrate nearly 2 million naturally and synthetically-derived compounds into the global network of biochemical knowledge, named ATLASx. ATLASx is available to researchers as a powerful online platform that supports the prediction and analysis of biochemical pathways and evaluates the biochemical vicinity of molecule classes (https://lcsb-databases.epfl.ch/Atlas2).
- Open Access
ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds
Synthetic biology and metabolic engineering rely on computational search tools for predictions of novel biosynthetic pathways to industrially important compounds, many of which are derived from aromatic amino acids. Pathway search tools vary in their scope of covered reactions and compounds, as well as in metrics for ranking and evaluation. In this work, we present a new computational resource called ARBRE: Aromatic compounds RetroBiosynthesis Repository and Explorer. It consists of a comprehensive biochemical reaction network centered around aromatic amino acid biosynthesis and a computational toolbox for navigating this network. ARBRE encompasses over 28’000 known and 100’000 novel reactions predicted with generalized enzymatic reactions rules and over 70’000 compounds, of which 22’000 are known to biochemical databases and 48’000 only to PubChem. Over 1,000 molecules that were solely part of the PubChem database before and were previously impossible to integrate into a biochemical network are included into the ARBRE reaction network by assigning enzymatic reactions. ARBRE can be applied for pathway search, enzyme annotation, pathway ranking, visualization, and network expansion around known biochemical pathways to predict valuable compound derivations. In line with the standards of open science, we have made the toolbox freely available to the scientific community at http://lcsb-databases.epfl.ch/arbre/ . We envision that ARBRE will provide the community with a new computational toolbox and comprehensive search tool to predict and rank pathways towards industrially important aromatic compounds.
Open AccessARBRE: Computational resource to predict pathways towards industrially important aromatic compounds
Synthetic biology and metabolic engineering rely on computational search tools for predictions of novel biosynthetic pathways to industrially important compounds, many of which are derived from aromatic amino acids. Pathway search tools vary in their scope of covered reactions and compounds, as well as in metrics for ranking and evaluation. In this work, we present a new computational resource called ARBRE: Aromatic compounds RetroBiosynthesis Repository and Explorer. It consists of a comprehensive biochemical reaction network centered around aromatic amino acid biosynthesis and a computational toolbox for navigating this network. ARBRE encompasses over 28’000 known and 100’000 novel reactions predicted with generalized enzymatic reactions rules and over 70’000 compounds, of which 22’000 are known to biochemical databases and 48’000 only to PubChem. Over 1,000 molecules that were solely part of the PubChem database before and were previously impossible to integrate into a biochemical network are included into the ARBRE reaction network by assigning enzymatic reactions. ARBRE can be applied for pathway search, enzyme annotation, pathway ranking, visualization, and network expansion around known biochemical pathways to predict valuable compound derivations. In line with the standards of open science, we have made the toolbox freely available to the scientific community at http://lcsb-databases.epfl.ch/arbre/ . We envision that ARBRE will provide the community with a new computational toolbox and comprehensive search tool to predict and rank pathways towards industrially important aromatic compounds.