Brain cell finder

bcfind is a tool for fully automated localization of soma in 3D mouse brain images acquired by confocal light sheet microscopy. The core technique in this method is supervised semantic deconvolution which uses a neural network to map a 3D image into a synthetic image where the visibility of specific entities of interest in the image (neural somata in this case) is enhanced and standardized.

Website: http://bcfind.dinfo.unifi.it/.

Related publications

  1. Ludovico Silvestri, Marco Paciscopi, Paolo Soda, Filippo Biamonte, Giulio Iannello, Paolo Frasconi, Francesco S Pavone (2015). Quantitative neuroanatomy of all Purkinje cells with light sheet microscopy and high-throughput image analysis. Frontiers in Neuroanatomy 9(68) Abstract
    Keywords: Brain; bcfind;
  2. Ludovico Silvestri, Nikita Rudinskiy, Marco Paciscopi, Irene Costantini, Leonardo Sacconi, Paolo Frasconi, Bradley T Hyman, Francesco S Pavone (2015). Optical mapping of neuronal activity with cellular resolution on a brain-wide scale. In SPIE European Conference on Biomedical Optics Abstract
    Keywords: Brain; Neuronal activity; Semantic deconvolution; Bioinformatics; bcfind;
  3. Alessandro Bria, Giulio Iannello, Paolo Soda, Hanchuan Peng, Giovanni Erbacci, Giuseppe Fiameni, G Mariani, R Mucci, M Rorro, Francesco S Pavone, Ludovico Silvestri, Paolo Frasconi, Roberto Cortini (2014). A HPC infrastructure for processing and visualizing neuro-anatomical images obtained by Confocal Light Sheet Microscopy. In High Performance Computing Simulation (HPCS), 2014 International Conference on (pp. 592–599) Abstract
    Keywords: Brain; Semantic deconvolution; bcfind; Bioinformatics;
  4. Paolo Frasconi, Ludovico Silvestri, Paolo Soda, Roberto Cortini, Francesco S Pavone, Giulio Iannello (2014). Large-Scale Automated Identification of Mouse Brain Cells in Confocal Light Sheet Microscopy Images. Bioinformatics 30(17):i587–i593 Abstract
    Keywords: Brain; Semantic deconvolution; bcfind; Bioinformatics;
 

 

Type extension trees

Type Extension Trees are a powerful representation language for describing count-of-count features in relational domains. These features characterize the combinatorial structure of relational neighborohoods, are more powerful than simple count statistics, and can be used to develop relational learning algorithms. In our new AIJ paper on type extension trees we present a structure learning algorithm for constructing TET from data, and take advantage of the Earth Mover's Distance to develop a supervised instance-based relational learner. We report experiments on bibliographic data show that TET-learning is able to discover the count-of-count feature underlying the definition of the h-index, and the inverse document frequency feature commonly used in information retrieval.

Related publications

  1. Manfred Jaeger, Marco Lippi, Andrea Passerini, Paolo Frasconi (2013). Type extension trees for feature construction and learning in relational domains. Artificial Intelligence 204:30–55 Abstract
    Keywords: Relational learning; Type extension trees; Feature learning;
  2. Paolo Frasconi, Manfred Jaeger, Andrea Passerini (2008). Feature Discovery with Type Extension Trees. In Filip Zelezny, Nada Lavrac (eds.), 18th International Conference on Inductive Logic Programming (ILP'08) (pp. 122–139) Abstract
    Keywords: Relational learning; Type extension trees; Feature learning;
 

 

kLog

kLog is a logical and relational language for kernel-based learning embedded in Prolog. It allows users to specify logical and relational learning problems at a high level, in a declarative way. It builds on simple but powerful concepts: learning from interpretations, entity/relationship data modeling, logic programming and deductive databases, and graph kernels. kLog can use numerical and symbolic data, background knowledge in the form of Prolog or Datalog programs (as in inductive logic programming systems) and several statistical procedures can be used to fit the model parameters.

Website: http://klog.dinfo.unifi.it/

.

Related publications

  1. Paolo Frasconi, Fabrizio Costa, Luc De Raedt, Kurt De Grave (2015). kLog: A Language for Logical and Relational Learning with Kernels (extended abstract). In Proc. International Joint Conference on Artificial Intelligence Journal Track.
    Keywords: Relational learning; kLog; Kernels; Graph Kernels;
  2. Paolo Frasconi, Fabrizio Costa, Luc De Raedt, Kurt De Grave (2014). kLog: A Language for Logical and Relational Learning with Kernels. Artificial Intelligence 217:117–143 Abstract
    Keywords: Relational learning; kLog; Kernels; Graph Kernels;
  3. Mathias Verbeke, Paolo Frasconi, Kurt De Grave, Fabrizio Costa, Luc De Raedt (2014). kLogNLP: Graph kernel--based relational learning of natural language. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics: System Demonstrations Abstract
    Keywords: Relational learning; Natural language processing; kLog;
  4. Laura Antanas, McElory Hoffmann, Paolo Frasconi, Tinne Tuytelaars, Luc De Raedt (2013). A relational kernel-based approach to scene classification. IEEE Workshop on Applications of Computer Vision 0:133–139 In None (pp. 133–139) Abstract
    Keywords: Relational learning; Computer vision; kLog;
  5. Mathias Verbeke, Vincent Van Asch, Roser Morante, Paolo Frasconi, Walter Daelemans, Luc De Raedt (2012). A Statistical Relational Learning Approach to Identifying Evidence Based Medicine Categories. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL) (pp. 579–589) Abstract
    Keywords: Relational learning; Natural language processing; kLog;
  6. Parisa Kordjamshidi, Paolo Frasconi, Martijn Van Otterlo, Marie-Francine Moens, Luc De Raedt (2012). Spatial Relation Extraction using Relational Learning. In Stephen Muggleton, Alireza Tamaddoni-Nezhad, Francesca A Lisi (eds.), 21st International Conference on Inductive Logic Programming (ILP'11) Lecture Notes in Computer Science (pp. 204–220) Abstract
    Keywords: Relational learning; Natural language processing; kLog;
  7. Mathias Verbeke, Paolo Frasconi, Vincent Van Asch, Roser Morante, Walter Daelemans, Luc De Raedt (2012). Kernel-based Logical and Relational Learning with kLog for Hedge Cue Detection. In Stephen Muggleton, Alireza Tamaddoni-Nezhad, Francesca A Lisi (eds.), 21st International Conference on Inductive Logic Programming (ILP'11) Lecture Notes in Computer Science (pp. 347–357) Abstract
    Keywords: Relational learning; Natural language processing; kLog;
  8. Mathias Verbeke, Paolo Frasconi, Vincent Van Asch, Roser Morante, Walter Daelemans, Luc De Raedt (2012). Kernel-based Logical and Relational Learning with kLog for Hedge Cue Detection. In Meeting of Computational Linguistics in The Netherlands (CLIN)
    Keywords: Relational learning; Natural language processing; kLog;
  9. Mathias Verbeke, Roser Morante, Paolo Frasconi, Luc De Raedt (2012). A statistical relational learning approach to identify sections in scientific abstracts using sentence and document structure. In Proceedings of the 21st Belgian-Dutch Conference on Machine Learning (BeNeLearn 2012)
    Keywords: Relational learning; Natural language processing; kLog;
  10. Laura Antanas, Paolo Frasconi, Fabrizio Costa, Tinne Tuytelaars, Luc De Raedt (2012). A Relational Kernel-Based Framework for Hierarchical Image Understanding. In Georgy L Gimel'farb, Edwin R Hancock, Atsushi Imiya, Arjan Kuijper, Mineichi Kudo, Shinichiro Omachi, Terry Windeatt, Keiji Yamada (eds.), Proc. of the Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition (SSPR{\&}SPR 2012) Lecture Notes in Computer Science (pp. 171–180) Abstract
    Keywords: Relational learning; Computer vision; kLog;
  11. Laura Antanas, Paolo Frasconi, Tinne Tuytelaars, Luc De Raedt (2011). Employing logical languages for image understanding. IEEE Workshop on Kernels and Distances for Computer Vision, International Conference on Computer Vision, Barcelona, Spain, 13 November 2011.
    Keywords: Relational learning; Computer vision; kLog;
 

Relational learning

Besides the two systems described above, we have developed other algorithms for learning in relational domains. Relational information gain is a novel refinement scoring function measuring the informativeness of newly introduced variables in ILP systems. We have also integrated kernel methods in relational learning systems in various way. kFOIL is a a dynamic propositionalization technique which uses a logical kernel and constructs clauses by leveraging FOIL search and using SVM performance to guide the search. Kernels on Prolog proof trees are based derived from background knowledge expressed in first-order logic.

Related publications

  1. Marco Lippi, Paolo Frasconi, Andrea Passerini (2012). Predicting Metal-Binding Sites from Protein Sequence. IEEE/ACM Transactions on Computational Biology and Bioinformatics 9(1):203–213 Abstract
    Keywords: Metal binding; Learning with structured outputs; Relational learning; Bioinformatics;
  2. Andrea Passerini, Marco Lippi, Paolo Frasconi (2011). MetalDetector v2.0: predicting the geometry of metal binding sites from protein sequence. Nucleic Acids Research 39(suppl 2):W288 Abstract
    Keywords: Metal binding; Learning with structured outputs; Relational learning; Bioinformatics;
  3. Marco Lippi, Manfred Jaeger, Paolo Frasconi, Andrea Passerini (2011). Relational information gain. Machine Learning 83(2):219–239 Abstract
    Keywords: Inductive logic programming; Relational learning;
  4. Marco Lippi, Matteo Bertini, Paolo Frasconi (2010). Collective Traffic Forecasting. In Jose Balcazar, Francesco Bonchi, Aristides Gionis, Michele Sebag (eds.), Machine Learning and Knowledge Discovery in Databases Lecture Notes in Computer Science (pp. 259–273) Abstract
    Keywords: Relational learning; Intelligent transportation systems; Traffic forecasting; Markov logic;
  5. Niels Landwehr, Andrea Passerini, Luc De Raedt, Paolo Frasconi (2010). Fast Learning of Relational Kernels. Machine Learning 78(3):305–342
    Keywords: Kernels; Inductive logic programming; Relational learning;
  6. Marco Lippi, Manfred Jaeger, Paolo Frasconi, Andrea Passerini (2009). Relational Information Gain (ILP). In Luc De Raedt (editor), 19th International Conference on Inductive Logic Programming (ILP'09) Abstract
    Keywords: Inductive logic programming; Relational learning; Relational features;
  7. Marco Lippi, Paolo Frasconi (2009). Prediction of Protein Beta-Residue Contacts by Markov Logic Networks with Grounding Specific Weights. Bioinformatics Abstract
    Keywords: Markov logic; Relational learning; Bioinformatics; Protein structure; Beta-sheets;
  8. Paolo Frasconi, Andrea Passerini (2008). Learning with Kernels and Logical Representations. In Luc De Raedt, Paolo Frasconi, Kristian Kersting, Stephen Muggleton (eds.), Probabilistic Inductive Logic Programming --- Theory and Applications LNAI 4911 (pp. 56–91) Abstract
    Keywords: Kernels; Inductive logic programming; Relational learning;
  9. Marco Lippi, Paolo Frasconi (2008). Markov Logic Improves Protein β-Partners Prediction. In 6th International Workshop on Mining and Learning with Graphs
    Keywords: Markov logic; Relational learning; Bioinformatics; Protein structure; Beta-sheets;
  10. Fabrizio Costa, Sauro Menchetti, Paolo Frasconi (2007). Comparing Sequence Classification Algorithms for Protein Subcellular Localization. In Barbara Hammer, Pascal Hitzler (eds.), Perspectives of Neural-Symbolic Integration (pp. 23–48) Abstract
    Keywords: Kernels; Sequence learning; Relational learning; Bioinformatics;
  11. Paolo Frasconi (2007). Learning with Kernels and Logical Representations (extended abstract). In H Blockeel, J Ramon, J Shavlik, P Tadepalli (eds.), Proc. 17th International Conference on Inductive Logic Programming LNAI 4893 (pp. 1–3) Extended abstract, invited keynote talk. Abstract
    Keywords: Kernels; Inductive logic programming; Relational learning;
  12. Niels Landwehr, Andrea Passerini, Luc De Raedt, Paolo Frasconi (2006). kFOIL: Learning Simple Relational Kernels. In Y Gil, R Mooney (eds.), Proc. Twenty-First National Conference on Artificial Intelligence (AAAI-06) Abstract
    Keywords: Kernels; Inductive logic programming; Relational learning;
  13. Andrea Passerini, Paolo Frasconi, Luc De Raedt (2006). Kernels on Prolog Proof Trees: Statistical Learning in the {ILP} Setting. Journal of Machine Learning Research 7:307–342 Abstract
    Keywords: Kernels; Inductive logic programming; Background knowledge; Relational learning;
  14. Andrea Passerini, Paolo Frasconi (2005). Kernels on Prolog Ground Terms. In Int. Joint Conf. on Artificial Intelligence (IJCAI'05) Abstract
    Keywords: Kernels; Inductive logic programming; Relational learning;
  15. Paolo Frasconi, Marco Gori, Andreas Kuechler, Alessandro Sperduti (2001). From Sequences to Data Structures: Theory and Applications. In J Kolen, S Kremer (eds.), A Field Guide to Dynamic Recurrent Networks (pp. 351–374)
    Keywords: Recursive neural networks; Relational learning;
  16. Paolo Frasconi (1998). An introduction to learning structured information. In C Lee Giles, Marco Gori (eds.), Adaptive Processing of Sequences and Data Structures Lecture Notes in Computer Science (pp. 99–120)
    Keywords: Recursive neural networks; Relational learning;
 

METALDETECTOR

MetalDetector identifies cysteines and histidines involved in transition metal protein binding sites, starting from the protein sequence alone. The prediction server is available at metaldetector.dsi.unifi.it.

Source code is also available.

Related publications

  1. Marco Lippi, Andrea Passerini, Marco Punta, Paolo Frasconi (2012). Metal Binding in Proteins: Machine Learning Complements X-Ray Absorption Spectroscopy. In Peter Flach, Tijl Bie, Nello Cristianini (eds.), Proc. of the European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2012) Lecture Notes in Computer Science (pp. 854–857) NECTAR-track. Abstract
    Keywords: Bioinformatics; Metal binding;
  2. Marco Lippi, Paolo Frasconi, Andrea Passerini (2012). Predicting Metal-Binding Sites from Protein Sequence. IEEE/ACM Transactions on Computational Biology and Bioinformatics 9(1):203–213 Abstract
    Keywords: Metal binding; Learning with structured outputs; Relational learning; Bioinformatics;
  3. Andrea Passerini, Marco Lippi, Paolo Frasconi (2011). MetalDetector v2.0: predicting the geometry of metal binding sites from protein sequence. Nucleic Acids Research 39(suppl 2):W288 Abstract
    Keywords: Metal binding; Learning with structured outputs; Relational learning; Bioinformatics;
  4. Wuxian Shi, Marco Punta, Jen Bohon, J Michael Sauder, Rhijuta D'Mello, Mike Sullivan, John Toomey, Don Abel, Marco Lippi, Andrea Passerini, Paolo Frasconi, Stephen K Burley, Burkhard Rost, and Mark R Chance (2011). Characterization of metalloproteins by high-throughput X-ray absorption spectroscopy. Genome Research 21(6):898–907 Abstract
    Keywords: Metal binding; Bioinformatics;
  5. Paolo Frasconi, Andrea Passerini (2009). Predicting the Geometry of Metal Binding Sites from Protein Sequence. In Daphne Koller, Dale Schuurmans, Yoshua Bengio, Léon Bottou (eds.), Proceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems (NIPS'08) (pp. 465–472) Abstract
    Keywords: Metal binding; Learning with structured outputs; Bioinformatics;
  6. Marco Lippi, Andrea Passerini, Marco Punta, Burkhard Rost, Paolo Frasconi (2008). MetalDetector: a web server for predicting metal binding sites and disulfide bridges in proteins from sequence. Bioinformatics 24(18):2094–2095 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
  7. Andrea Passerini, Claudia Andreini, Sauro Menchetti, Antonio Rosato, Paolo Frasconi (2007). Predicting zinc binding at the proteome level. BMC Bioinformatics 8(39) Abstract
    Keywords: Metal binding; Bioinformatics;
  8. Sauro Menchetti, Andrea Passerini, Paolo Frasconi, Claudia Andreini, Antonio Rosato (2006). Improving Prediction of Zinc Binding Sites by Modeling the Linkage between Residues Close in Sequence. In Alberto Apostolico, Concettina Guerra, Sorin Istrail, Pavel A Pevzner, Michael S Waterman (eds.), Proc. Tenth Annual International Conference on Research in Computational Molecular Biology (RECOMB'06) LNCS (pp. 309–320) Abstract
    Keywords: Metal binding; Bioinformatics;
  9. Andrea Passerini, Marco Punta, Alessio Ceroni, Burkhard Rost, Paolo Frasconi (2006). Identifying cysteines and histidines in transition-metal-binding sites using support vector machines and neural networks. Proteins: Structure, Function, and Bioinformatics (65):305–316 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
  10. Andrea Passerini, Paolo Frasconi (2004). Learning to discriminate between ligand bound and disulfide bound cysteines. Protein Engineering Design and Selection 7(4):367–73 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
 

DISULFIND

Disulfind predicts disulfide bridges in proteins, starting from protein sequence alone. The prediction server is available atdisulfind.dsi.unifi.it. If you are on Debian or Ubuntu you may install a standalone version:

sudo apt-get install disulfinder

disulfinder --help

Both Metaldetector and Disuldind are also integrated in PredictProtein.

Related publications

  1. Marco Lippi, Andrea Passerini, Marco Punta, Burkhard Rost, Paolo Frasconi (2008). MetalDetector: a web server for predicting metal binding sites and disulfide bridges in proteins from sequence. Bioinformatics 24(18):2094–2095 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
  2. Marc Vincent, Andrea Passerini, Mattheu Labbe, Paolo Frasconi (2008). A simplified approach to disulfide connectivity prediction from protein sequences. BMC Bioinformatics 9(20) Abstract
    Keywords: Disulfide bridges; Bioinformatics;
  3. Andrea Passerini, Marco Punta, Alessio Ceroni, Burkhard Rost, Paolo Frasconi (2006). Identifying cysteines and histidines in transition-metal-binding sites using support vector machines and neural networks. Proteins: Structure, Function, and Bioinformatics (65):305–316 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
  4. Alessio Ceroni, Andrea Passerini, Alessandro Vullo, Paolo Frasconi (2006). {DISULFIND}: a Disulfide Bonding State and Cysteine Connectivity Prediction Server. Nucleic Acids Research 34:W177–W181 Web Server Issue. Abstract
    Keywords: Disulfide bridges; Bioinformatics;
  5. Andrea Passerini, Paolo Frasconi (2004). Learning to discriminate between ligand bound and disulfide bound cysteines. Protein Engineering Design and Selection 7(4):367–73 Abstract
    Keywords: Metal binding; Disulfide bridges; Bioinformatics;
  6. Alessandro Vullo, Paolo Frasconi (2004). Disulfide Connectivity Prediction Using Recursive Neural Networks and Multiple Alignments. Bioinformatics 20(5):653–9 Abstract
    Keywords: Disulfide bridges; Learning with structured outputs; Recursive neural networks; Bioinformatics;
  7. Alessandro Vullo, Paolo Frasconi (2003). A Recursive Connectionist Approach for Predicting Disulfide Connectivity in Proteins. In Eighteenth Annual ACM Symposium on Applied Computing (SAC '03) (pp. 67–71) Special Track on Bioinformatics. Abstract
    Keywords: Disulfide bridges; Bioinformatics; Recursive neural networks; Learning with structured outputs;
  8. Alessio Ceroni, Paolo Frasconi, Andrea Passerini, Alessandro Vullo (2003). Predicting the Disulfide Bonding State of Cysteines with Combinations of Kernel Machines. Journal of VLSI SIgnal Processing 35(3):287–295 Abstract
    Keywords: Disulfide bridges; Bioinformatics;
  9. Paolo Frasconi, Andrea Passerini, Alessandro Vullo (2002). A Two-Stage {SVM} Architecture for Predicting the Disulfide Bonding State of Cysteines. In IEEE Neural Networks for Signal Processing (pp. 25–34) Abstract
    Keywords: Disulfide bridges; Bioinformatics;
 

 

MLOCSR

MLOCSR converts bitmap images of chemical structural formulae into machine readable vector formats (such as MOL and SDF).

Website: http://mlocsr.dinfo.unifi.it/.

Related publications

  1. Paolo Frasconi, Francesco Gabbrielli, Marco Lippi, Simone Marinai (2014). Markov Logic Networks for Optical Chemical Structure Recognition. Journal of Chemical Information and Modeling 54(8):2380–2390 In press. Abstract
    Keywords: MLOCSR; Chemoinformatics;
 

 

Small molecules

We have developed graph kernels predicting the activity of small molecules using 2D and 3D representations. Source code written by Fabrizio Costa and Alessio Ceroni for the original method (which includes a 3D kernel) is available here. For the 2D case a more recent and significantly evolved approach is EDeN developed at the University of Freiburg.

Related publications

  1. Alessio Ceroni, Fabrizio Costa, Paolo Frasconi (2007). Classification of small molecules by two-and three-dimensional decomposition kernels. Bioinformatics 23(16):2038–2045 Abstract
    Keywords: Kernels; Small molecules; Bioinformatics; Chemoinformatics;
  2. Sauro Menchetti, Fabrizio Costa, Paolo Frasconi (2005). Weighted Decomposition Kernels. In Luc De Raedt, Stefan Wrobel (eds.), Proc. Int. Conf. on Machine Learning (ICML'05) Abstract
    Keywords: Kernels; Graph kernels; Sequence learning; Bioinformatics; Small molecules; Chemoinformatics;
 

 

Protein structure prediction

We have applied several machine learning methods to the prediction of protein structure. Source code of a Beta-residue contacts predictor, based on neural networks and Markov Logic, (written by Marco Lippi) is available for download. The bidirectional RNN described in Baldi et al (1999) was originally developed by G. Pollastri in our lab and has subsequently evolved into the state-of-the-art systems SSpro and PORTER.

Related publications

  1. Marco Lippi, Paolo Frasconi (2009). Prediction of Protein Beta-Residue Contacts by Markov Logic Networks with Grounding Specific Weights. Bioinformatics Abstract
    Keywords: Markov logic; Relational learning; Bioinformatics; Protein structure; Beta-sheets;
  2. Alessandro Vullo, Andrea Passerini, Paolo Frasconi, Fabrizio Costa, Gianluca Pollastri (2008). On the Convergence of Protein Structure and Dynamics. Statistical Learning Studies of Pseudo Folding Pathways. In E Marchiori, J H Moore (eds.), Proc. 6th European Conf. on Evolutionary Computation, Machine Learning and Data Mining in Bioinformatics (EVOBIO'08) LNCS 4973 (pp. 200–211) Abstract
    Keywords: Protein structure; Bioinformatics;
  3. Marco Lippi, Paolo Frasconi (2008). Markov Logic Improves Protein β-Partners Prediction. In 6th International Workshop on Mining and Learning with Graphs
    Keywords: Markov logic; Relational learning; Bioinformatics; Protein structure; Beta-sheets;
  4. Gianluca Pollastri, Alessandro Vullo, Paolo Frasconi, Pierre Baldi (2006). Modular DAG-RNN Architectures for Assembling Coarse Protein Structures. Journal of Computational Biology 13(3):631–650 Abstract
    Keywords: Protein structure; Contact maps; Recursive neural networks; Bioinformatics;
  5. Alessio Ceroni, Paolo Frasconi, Alessandro Vullo (2005). Protein Structure Assembly from Knowledge of Beta-Sheet Motifs and Secondary Structure. In Bruno Apolloni, Maria Marinaro, Roberto Tagliaferri (eds.), Biological and Artificial Intelligence Environments (pp. 45–52) Abstract
    Keywords: Protein structure; Beta-sheets; Recursive neural networks;
  6. Alessio Ceroni, Paolo Frasconi, Gianluca Pollastri (2005). Learning Protein Secondary Structure from Sequential and Relational Data. Neural Networks 18(8):1029–39 Abstract
    Keywords: Protein structure; Recurrent neural networks;
  7. Pierre Baldi, Gianluca Pollastri, Paolo Frasconi, Alessandro Vullo (2003). New Machine Learning Methods for the Prediction of Protein Topologies. In Paolo Frasconi, R Shamir (eds.), Artificial Intelligence and Heuristic Methods for Bioinformatics
    Keywords: Protein structure; Bioinformatics;
  8. Alessandro Vullo, Paolo Frasconi (2003). Prediction of Protein Coarse Contact Maps. Journal of Bioinformatics and Computational Biology 1(2):411–431 Abstract
    Keywords: Protein structure; Contact maps; Recursive neural networks; Bioinformatics;
  9. Gianluca Pollastri, Pierre Baldi, Alessandro Vullo, Paolo Frasconi (2002). Prediction of Protein Topologies Using Generalized {IOHMMs} and {RNNs}. In S Becker, S Thrun, K Obermayer (eds.), Neural Information Processing Systems (NIPS'02) (pp. 1449–1456) Abstract
    Keywords: Protein structure; Contact maps; Recursive neural networks; Bioinformatics;
  10. Alessandro Vullo, Paolo Frasconi (2002). A Bi-Recursive Neural Network Architecture for the Prediction of Protein Coarse Contact Maps. In 1st IEEE Computer Society Bioinformatics Conference (CSB'02) (pp. 187–196)
    Keywords: Protein structure; Contact maps; Recursive neural networks; Learning with structured outputs; Bioinformatics;
  11. Paolo Frasconi, Alessandro Vullo (2002). Prediction of Protein Coarse Contact Maps using Recursive Neural Networks. In Proc. IEEE-EMBS Conference on Molecular, Cellular, and Tissue Engineering
    Keywords: Protein structure; Contact maps; Recursive neural networks; Learning with structured outputs; Bioinformatics;
  12. Soeren Brunak, Pierre Baldi, Paolo Frasconi, Gianluca Pollastri, Giovanni Soda (2001). Bidirectional {IOHMMs} and Recurrent Neural Netoworks for Protein Secondary Structure Prediction. In R Casadio, L Masotti (eds.), In Protein Sequence Analysis In The Post-Genomic Era
    Keywords: Recurrent neural networks; Bioinformatics; Protein structure;
  13. Pierre Baldi, Soeren Brunak, Paolo Frasconi, Gianluca Pollastri, Giovanni Soda (2001). Bidirectional Dynamics for Protein Secondary Structure Prediction. In Ron Sun, C Lee Giles (eds.), In Sequence Learning: Paradigms, Algorithms, and Applications Lecture Notes in Computer Science (pp. 80–104) Abstract
    Keywords: Recurrent neural networks; Bioinformatics; Protein structure;
  14. Pierre Baldi, Soeren Brunak, Paolo Frasconi, Giovanni Soda, Gianluca Pollastri (1999). Exploiting the past and the future in protein secondary structure prediction. Bioinformatics 15(11):937–946 Abstract
    Keywords: Protein structure; Recurrent neural networks; Bioinformatics;