Pathway Analyst Automated Metabolic Pathway Prediction

Pireddu, Luca; Poulin, Brett; Szafron, Duane; Lu, Philip; Wishart, David S.

doi:10.1109/cibcb.2005.1594924

Cited by 11 publications

(11 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This experience suggests that if the PLR iterations converge too quickly the estimates of b m reach a local optimum. A subsequent effect is the HME likelihood in the following iterations becomes erratic as the EM responsibilities (4) are dominated by the PLR probabilities p(y|x, b m ) which do not necessarily reflect the structure within the 3M parameters. The different rates of convergence between the 3M and PLR parameters can cause instabilities in the HME3M likelihood.…”

Section: Penalized Logistic Regression (Plr)mentioning

confidence: 99%

A Markov Classification Model for Metabolic Pathways

Hancock

Mamitsuka

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Background: This paper considers the problem of identifying pathways through metabolic networks that relate to a specific biological response. Our proposed model, HME3M, first identifies frequently traversed network paths using a Markov mixture model. Then by employing a hierarchical mixture of experts, separate classifiers are built using information specific to each path and combined into an ensemble prediction for the response. Results: We compared the performance of HME3M with logistic regression and support vector machines (SVM) for both simulated pathways and on two metabolic networks, glycolysis and the pentose phosphate pathway for Arabidopsis thaliana. We use AltGenExpress microarray data and focus on the pathway differences in the developmental stages and stress responses of Arabidopsis. The results clearly show that HME3M outperformed the comparison methods in the presence of increasing network complexity and pathway noise. Furthermore an analysis of the paths identified by HME3M for each metabolic network confirmed known biological responses of Arabidopsis.Conclusions: This paper clearly shows HME3M to be an accurate and robust method for classifying metabolic pathways. HME3M is shown to outperform all comparison methods and further is capable of identifying known biologically active pathways within microarray data.

show abstract

Section: Penalized Logistic Regression (Plr)mentioning

confidence: 99%

A Markov Classification Model for Metabolic Pathways

Hancock

Mamitsuka

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Path-A uses a set of machine-learning techniques to predict pathway instances ( 5 ). The basic prediction algorithm is based on the realization that instances of the same pathway in different organisms will share some degree of similarity, in both the reactions they encompass and the primary sequence structure of the proteins that catalyse them.…”

Section: Prediction Techniquementioning

confidence: 99%

“…We have evaluated several classifiers as predictors of which proteins are functionally compatible ( 5 ). A key research result, established using Path-A, is that no single classifier is best for all pathway reactions.…”

Section: Prediction Techniquementioning

confidence: 99%

The Path-A metabolic pathway prediction web server

Pireddu

Szafron

et al. 2006

Nucleic Acids Research

Self Cite

View full text Add to dashboard Cite

Pathway Analyst (Path-A) is a publicly available web server () that predicts metabolic pathways. It takes a FASTA format file containing a set of query protein sequences from a single organism (a partial or complete proteome) and identifies those sequences that are likely to participate in any of its supported metabolic pathways (currently 10). Path-A uses a number of machine-learning and sequence analysis techniques (e.g. SVM, BLAST and HMM) to predict pathways. Each machine-learned classifier exploits similarity between sequences in the pathways of its model organisms and sequences in the query set. It predicts the pathways that are present in the query organism and annotates each predicted reaction and catalyst, using the appropriate sequences from the query set. Path-A also provides a browsable and searchable database of the pathways for the model organisms that are used to make its predictions. Path-A's predictor sets (using different classifier technologies) have been evaluated using standard cross-validation techniques on a dataset of 10 metabolic pathways across 13 model organisms—a total of 125 organism-specific pathways. The most accurate classifier technology obtained a mean precision of 78.3% and a mean recall of 92.6% in predicting all catalyst proteins, of all reactions, in all pathways present in the dataset. Although Path-A currently only supports metabolic pathways, the underlying prediction techniques are general enough for other types of pathways. Consequently, it is our intent to extend Path-A to predict other types of pathways, including signalling pathways.

show abstract

“…Protein sequences are taken as a single FASTA file from a single organism (complete or a partial proteome) and it identifies the sequences which participate in the metabolic pathways available in the server (Pireddu et al 2005(Pireddu et al , 2006.…”

Section: Pathway Databasesmentioning

confidence: 99%

BioSilicoSystems - A Multipronged Approach Towards Analysis and Representation of Biological Data (PhD Thesis)

Hariharaputran

2010

Nat Prec

View full text Add to dashboard Cite

The rising field of integrative bioinformatics provides the vital methods to integrate, manage and also to analyze the diverse data and allows gaining new and deeper insights and a clear understanding of the intricate biological systems. The difficulty is not only to facilitate the study of heterogeneous data within the biological context, but it also more fundamental, how to represent and make the available knowledge accessible. Moreover, adding valuable information and functions that persuade the user to discover the interesting relations hidden within the data is, in itself, a great challenge. Also, the cumulative information can provide greater biological insight than is possible with individual information sources. Furthermore, the rapidly growing number of databases and data types poses the challenge of integrating the heterogeneous data types, especially in biology. This rapid increase in the volume and number of data resources drive for providing polymorphic views of the same data and often overlap in multiple resources.In this thesis a multi-pronged approach is proposed that deals with various methods for the analysis and representation of the diverse biological data which are present in different data sources. This is an effort to explain and emphasize on different concepts which are developed for the analysis of molecular data and also to explain its biological significance. The hypotheses proposed are in context with various other results and findings published in the past. The approach demonstrated also explains different ways to integrate the molecular data from various sources along with the need for a comprehensive understanding and clear projection of the concept or the algorithm and its results, but with simple means and methods. The multifarious approach proposed in this work comprises of different tools or methods spanning significant areas of bioinformatics research such as data integration, data visualization, biological network construction / reconstruction and alignment of biological pathways. Each tool deals with a unique approach to utilize the molecular data for different areas of biological research and is built based on the kernel of the thesis. Furthermore these methods are combined with graphical representation that make things simple and comprehensible and also helps to understand with ease the underlying biological complexity. Moreover the human eye is often used to and it is more comfortable with the visual representation of the facts.

show abstract

Pathway Analyst Automated Metabolic Pathway Prediction

Cited by 11 publications

References 8 publications

A Markov Classification Model for Metabolic Pathways

A Markov Classification Model for Metabolic Pathways

The Path-A metabolic pathway prediction web server

BioSilicoSystems - A Multipronged Approach Towards Analysis and Representation of Biological Data (PhD Thesis)

Contact Info

Product

Resources

About