Prediction, expansion, and visualization of biological pathways and networks using perturbation data and cyclical graphical models

نویسنده

  • Charles Vaske
چکیده

Cellular processes are the interaction of multiple proteins, genomic sites, RNAs, small molecules, and their complexes. The set of these interactions and their contexts provide biological understanding of functionality beyond single-gene annotation. Biological networks have emerged as the dominant method of communicating, modeling, and understanding cellular processes and pathways. Computational prediction of interactions and networks is an open problem under active research. New high-throughput experimental techniques for measuring and perturbing gene expression, detecting proteinprotein interactions, and detecting protein-DNA interactions are providing rich datasets for predictions. Current methods are able to accurately predict some known pathways using of these data types. However, current methods have some limitations and are not able to fully analyze current datasets. For example, current models cannot learn negative feedback, a common network motif. Limited models of gene regulation also fail to find additive interactions. Also, current learning methods do not explicitly provide biologists guidance on follow-up experiments. I propose a new computational framework using factor graphs to address these limitations. The framework can model signed cycles of regulation, can incorporate multiple-gene knockdowns to infer additive interactions, and suggests follow-up experiments based on an information theoretic active learning approach. Preliminary results show the ability to recover signaling networks using gene-expression effects from gene knockdowns. Finally, I propose a new web tool for visualizing networks. This tool will be able to confirm network predictions through comparison with independent high-throughput datasets. The web tool will also allow easy collaboration and sharing of network predictions and network layouts.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Modeling gene regulatory networks: Classical models, optimal perturbation for identification of network

Deep understanding of molecular biology has allowed emergence of new technologies like DNA decryption.  On the other hand, advancements of molecular biology have made manipulation of genetic systems simpler than ever; this promises extraordinary progress in biological, medical and biotechnological applications.  This is not an unrealistic goal since genes which are regulated by gene regulatory ...

متن کامل

Comparison of Artificial Neural Networks and Cox Regression Models in Prediction of Kidney Transplant Survival

Cox regression model serves as a statistical method for analyzing the survival data, which requires some options such as hazard proportionality. In recent decades, artificial neural network model has been increasingly applied to predict survival data. This research was conducted to compare Cox regression and artificial neural network models in prediction of kidney transplant survival. The prese...

متن کامل

Comparison of Artificial Neural Networks and Cox Regression Models in Prediction of Kidney Transplant Survival

Cox regression model serves as a statistical method for analyzing the survival data, which requires some options such as hazard proportionality. In recent decades, artificial neural network model has been increasingly applied to predict survival data. This research was conducted to compare Cox regression and artificial neural network models in prediction of kidney transplant survival. The prese...

متن کامل

Prediction of Blasting Cost in Limestone Mines Using Gene Expression Programming Model and Artificial Neural Networks

The use of blasting cost (BC) prediction to achieve optimal fragmentation is necessary in order to control the adverse consequences of blasting such as fly rock, ground vibration, and air blast in open-pit mines. In this research work, BC is predicted through collecting 146 blasting data from six limestone mines in Iran using the artificial neural networks (ANNs), gene expression programming (G...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007