Ivan Rossi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ivan Rossi has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Piero Fariselli
AI and Computational Biomedicine Unit, Department of Medical Sciences, University of Turin, Turin, Italy
9 shared publications - Tiziana Sanavia
AI and Computational Biomedicine Unit, Department of Medical Sciences, University of Turin, Turin, Italy
6 shared publications - Cesare Rollo
Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, Torino, Italy
5 shared publications - Giovanni Birolo
University of Turin
4 shared publications - Corrado Pancotti
AI and Computational Biomedicine Unit, Department of Medical Sciences, University of Turin, Turin, Italy
3 shared publications - Pier Luigi Martelli
Biocomputing Group, University of Bologna, 40126 Bologna, Italy
3 shared publications
Latest Publications
- Open Access
- 3 February 2026
- Published by Springer Nature in Communications Biology
- Vol. 9 (1)
JanusDDG: a physics-informed neural network for sequence-based protein stability via two-fronts attention
Predicting how residue variations affect protein stability is crucial for rational protein design and for assessing the impact of disease-related mutations. Recent advances in protein language models have revolutionized computational protein analysis, enabling more accurate predictions of mutational effects. However, balancing predictive accuracy with the fundamental laws of thermodynamics remains a challenge for sequence-based models. Here we show JanusDDG, a physics-informed neural network that leverages embeddings from protein language models and a bidirectional cross-attention transformer architecture to predict stability changes for both single and multiple residue mutations. By adopting a physics-informed paradigm, the model is explicitly constrained to satisfy fundamental thermodynamic principles, such as antisymmetry and transitivity, while maintaining high predictive performance. Instead of conventional self-attention, JanusDDG employs a cross-interleaved attention mechanism that computes the relationship between wild-type and mutant embeddings to capture mutation-induced perturbations while preserving essential contextual information. Our results demonstrate that JanusDDG achieves state-of-the-art performance in predicting stability changes from sequence alone, matching or exceeding the accuracy of structure-based methods for both single and multiple mutations.
- 3 February 2026
- Published by Springer Nature in Communications Biology
- Vol. 9 (1)
Open AccessJanusDDG: a physics-informed neural network for sequence-based protein stability via two-fronts attention
Predicting how residue variations affect protein stability is crucial for rational protein design and for assessing the impact of disease-related mutations. Recent advances in protein language models have revolutionized computational protein analysis, enabling more accurate predictions of mutational effects. However, balancing predictive accuracy with the fundamental laws of thermodynamics remains a challenge for sequence-based models. Here we show JanusDDG, a physics-informed neural network that leverages embeddings from protein language models and a bidirectional cross-attention transformer architecture to predict stability changes for both single and multiple residue mutations. By adopting a physics-informed paradigm, the model is explicitly constrained to satisfy fundamental thermodynamic principles, such as antisymmetry and transitivity, while maintaining high predictive performance. Instead of conventional self-attention, JanusDDG employs a cross-interleaved attention mechanism that computes the relationship between wild-type and mutant embeddings to capture mutation-induced perturbations while preserving essential contextual information. Our results demonstrate that JanusDDG achieves state-of-the-art performance in predicting stability changes from sequence alone, matching or exceeding the accuracy of structure-based methods for both single and multiple mutations.
- Open Access
- 17 October 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 198 (Pt B) , 111176
Beyond Cox models: Assessing the performance of machine-learning methods in non-proportional hazards and non-linear survival analysis
- 17 October 2025
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 198 (Pt B) , 111176
Open AccessBeyond Cox models: Assessing the performance of machine-learning methods in non-proportional hazards and non-linear survival analysis
- Open Access
- 25 April 2025
- Published by Wiley in Protein Science
- Vol. 34 (5) , e70134
Mass balance approximation of unfolding boosts potential‐based protein stability predictions
Predicting protein stability changes upon single‐point mutations is crucial in computational biology, with applications in drug design, enzyme engineering, and understanding disease mechanisms. While deep‐learning approaches have emerged, many remain inaccessible for routine use. In contrast, potential‐like methods, including deep‐learning‐based ones, are faster, user‐friendly, and effective in estimating stability changes. However, most of them approximate Gibbs free‐energy differences without accounting for the free‐energy changes of the unfolded state, violating mass balance and potentially reducing accuracy. Here, we show that incorporating mass balance as a first approximation of the unfolded state significantly improves potential‐like methods. While many machine‐learning models implicitly or explicitly use mass balance, our findings suggest that a more accurate unfolded‐state representation could further enhance stability change predictions.
- 25 April 2025
- Published by Wiley in Protein Science
- Vol. 34 (5) , e70134
Open AccessMass balance approximation of unfolding boosts potential‐based protein stability predictions
Predicting protein stability changes upon single‐point mutations is crucial in computational biology, with applications in drug design, enzyme engineering, and understanding disease mechanisms. While deep‐learning approaches have emerged, many remain inaccessible for routine use. In contrast, potential‐like methods, including deep‐learning‐based ones, are faster, user‐friendly, and effective in estimating stability changes. However, most of them approximate Gibbs free‐energy differences without accounting for the free‐energy changes of the unfolded state, violating mass balance and potentially reducing accuracy. Here, we show that incorporating mass balance as a first approximation of the unfolded state significantly improves potential‐like methods. While many machine‐learning models implicitly or explicitly use mass balance, our findings suggest that a more accurate unfolded‐state representation could further enhance stability change predictions.
- Open Access
SurvHive: a package to consistently access multiple survival-analysis packages
Survival analysis, a foundational tool for modeling time-to-event data, has seen growing integration with machine learning (ML) approaches to handle the complexities of censored data and time-varying risks. Despite these advances, leveraging state-of-the-art survival models remains a challenge due to the fragmented nature of existing implementations, which lack standardized interfaces and require extensive preprocessing. We introduce SurvHive, a Python-based framework designed to unify survival analysis methods within a coherent and extensible interface modeled on scikit-learn. SurvHive integrates classical statistical models with cutting-edge deep learning approaches, including transformer-based architectures and parametric survival models. Using a consistent API, SurvHive simplifies model training, evaluation, and optimization, significantly reducing the barrier to entry for ML practitioners exploring survival analysis. The package includes enhanced support for hyper-parameter tuning, time-dependent risk evaluation metrics, and cross-validation strategies tailored to censored data. With its extensibility and focus on usability, SurvHive provides a bridge between survival analysis and the broader ML community, facilitating advancements in time-to-event modeling across domains. The SurvHive code and documentation are available freely at https://github.com/compbiomed-unito/survhive.
Open AccessSurvHive: a package to consistently access multiple survival-analysis packages
Survival analysis, a foundational tool for modeling time-to-event data, has seen growing integration with machine learning (ML) approaches to handle the complexities of censored data and time-varying risks. Despite these advances, leveraging state-of-the-art survival models remains a challenge due to the fragmented nature of existing implementations, which lack standardized interfaces and require extensive preprocessing. We introduce SurvHive, a Python-based framework designed to unify survival analysis methods within a coherent and extensible interface modeled on scikit-learn. SurvHive integrates classical statistical models with cutting-edge deep learning approaches, including transformer-based architectures and parametric survival models. Using a consistent API, SurvHive simplifies model training, evaluation, and optimization, significantly reducing the barrier to entry for ML practitioners exploring survival analysis. The package includes enhanced support for hyper-parameter tuning, time-dependent risk evaluation metrics, and cross-validation strategies tailored to censored data. With its extensibility and focus on usability, SurvHive provides a bridge between survival analysis and the broader ML community, facilitating advancements in time-to-event modeling across domains. The SurvHive code and documentation are available freely at https://github.com/compbiomed-unito/survhive.
- Open Access
- 15 March 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 172, 108288
SYNDSURV: A simple framework for survival analysis with data distributed across multiple institutions
- 15 March 2024
- Published by Elsevier in Computers in Biology and Medicine
- Vol. 172, 108288
Open AccessSYNDSURV: A simple framework for survival analysis with data distributed across multiple institutions
- Open Access
Influence of Model Structures on Predictors of Protein Stability Changes from Single-Point Mutations
Missense variation in genomes can affect protein structure stability and, in turn, the cell physiology behavior. Predicting the impact of those variations is relevant, and the best-performing computational tools exploit the protein structure information. However, most of the current protein sequence variants are unresolved, and comparative or ab initio tools can provide a structure. Here, we evaluate the impact of model structures, compared to experimental structures, on the predictors of protein stability changes upon single-point mutations, where no significant changes are expected between the original and the mutated structures. We show that there are substantial differences among the computational tools. Methods that rely on coarse-grained representation are less sensitive to the underlying protein structures. In contrast, tools that exploit more detailed molecular representations are sensible to structures generated from comparative modeling, even on single-residue substitutions.
Open AccessInfluence of Model Structures on Predictors of Protein Stability Changes from Single-Point Mutations
Missense variation in genomes can affect protein structure stability and, in turn, the cell physiology behavior. Predicting the impact of those variations is relevant, and the best-performing computational tools exploit the protein structure information. However, most of the current protein sequence variants are unresolved, and comparative or ab initio tools can provide a structure. Here, we evaluate the impact of model structures, compared to experimental structures, on the predictors of protein stability changes upon single-point mutations, where no significant changes are expected between the original and the mutated structures. We show that there are substantial differences among the computational tools. Methods that rely on coarse-grained representation are less sensitive to the underlying protein structures. In contrast, tools that exploit more detailed molecular representations are sensible to structures generated from comparative modeling, even on single-residue substitutions.
- 16 September 2023
- Published by Elsevier in Computer Physics Communications
- Vol. 294, 108933
Polyrate 2023: A computer program for the calculation of chemical reaction rates for polyatomics. New version announcement
- 16 September 2023
- Published by Elsevier in Computer Physics Communications
- Vol. 294, 108933
Polyrate 2023: A computer program for the calculation of chemical reaction rates for polyatomics. New version announcement
- Open Access
- 26 May 2011
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 39 (suppl) , W197-W202
BAR-PLUS: the Bologna Annotation Resource Plus for functional and structural annotation of protein sequences
We introduce BAR-PLUS (BAR + ), a web server for functional and structural annotation of protein sequences. BAR + is based on a large-scale genome cross comparison and a non-hierarchical clustering procedure characterized by a metric that ensures a reliable transfer of features within clusters. In this version, the method takes advantage of a large-scale pairwise sequence comparison of 13 495 736 protein chains also including 988 complete proteomes. Available sequence annotation is derived from UniProtKB, GO, Pfam and PDB. When PDB templates are present within a cluster (with or without their SCOP classification), profile Hidden Markov Models (HMMs) are computed on the basis of sequence to structure alignment and are cluster-associated (Cluster-HMM). Therefrom, a library of 10 858 HMMs is made available for aligning even distantly related sequences for structural modelling. The server also provides pairwise query sequence–structural target alignments computed from the correspondent Cluster-HMM. BAR + in its present version allows three main categories of annotation: PDB [with or without SCOP (*)] and GO and/or Pfam; PDB (*) without GO and/or Pfam; GO and/or Pfam without PDB (*) and no annotation. Each category can further comprise clusters where GO and Pfam functional annotations are or are not statistically significant. BAR + is available at http://bar.biocomp.unibo.it/bar2.0 .
- 26 May 2011
- Published by Oxford University Press (OUP) in Nucleic Acids Research
- Vol. 39 (suppl) , W197-W202
Open AccessBAR-PLUS: the Bologna Annotation Resource Plus for functional and structural annotation of protein sequences
We introduce BAR-PLUS (BAR + ), a web server for functional and structural annotation of protein sequences. BAR + is based on a large-scale genome cross comparison and a non-hierarchical clustering procedure characterized by a metric that ensures a reliable transfer of features within clusters. In this version, the method takes advantage of a large-scale pairwise sequence comparison of 13 495 736 protein chains also including 988 complete proteomes. Available sequence annotation is derived from UniProtKB, GO, Pfam and PDB. When PDB templates are present within a cluster (with or without their SCOP classification), profile Hidden Markov Models (HMMs) are computed on the basis of sequence to structure alignment and are cluster-associated (Cluster-HMM). Therefrom, a library of 10 858 HMMs is made available for aligning even distantly related sequences for structural modelling. The server also provides pairwise query sequence–structural target alignments computed from the correspondent Cluster-HMM. BAR + in its present version allows three main categories of annotation: PDB [with or without SCOP (*)] and GO and/or Pfam; PDB (*) without GO and/or Pfam; GO and/or Pfam without PDB (*) and no annotation. Each category can further comprise clusters where GO and Pfam functional annotations are or are not statistically significant. BAR + is available at http://bar.biocomp.unibo.it/bar2.0 .
- 1 January 2008
- Published by Springer Nature
- p. 309-326
The state of the art of membrane protein structure prediction: from sequence to 3D structure
- 1 January 2008
- Published by Springer Nature
- p. 309-326
The state of the art of membrane protein structure prediction: from sequence to 3D structure
- 1 January 2006
- Published by Elsevier
- p. 359-370
Machine learning and the prediction of protein structure: the state of the art
- 1 January 2006
- Published by Elsevier
- p. 359-370
Machine learning and the prediction of protein structure: the state of the art