Find results with:

Add another field

Search by date

to

Searching collections:

Dartmouth Dissertations

Add or remove collections

Home Dartmouth Dissertations jiang2016 1

Reference URL

To link to the entire object, paste this link in email, IM or document

To embed the entire object, paste this HTML in website

To link to this page, paste this link in email, IM or document

To embed this page, paste this HTML in website

Improving MHC-peptide binding prediction from sparse data with a multi-allele graphical model and active learning / by Li Jiang

you wish to report:

Your comment:

Your Name:

Submit

Cancel

...

Back to top

Powered by CONTENTdm®

Author	Jiang, Li
Title	Improving MHC-peptide binding prediction from sparse data with a multi-allele graphical model and active learning / by Li Jiang
Date	2016
Abstract	The recognition of foreign peptides by MHC (Major Histocompatibility Complex) proteins is a critical step in the immune response mounted by higher vertebrates against pathogens. MHC proteins have many allelic variants, each recognizing different peptides to different extents. Thus experimental determination of the binding specificity of each MHC variant is infeasible, and computational models are necessary to predict MHC-peptide binding. While many computational approaches have proved effective given sufficient data, they tend to break down in data-scarce scenarios. This thesis presents computational models and methods to enable prediction of MHC-peptide binding from limited data. First, a multi-allele probabilistic model is developed to exploit similarities among allele sequences and best leverage the limited experimental data available for the vast majority of alleles, including those from model organisms. Second, this model is employed in algorithms that optimize MHC-peptide binding experiments, selecting peptides to improve the model for a target allele. Cross-validation comparisons show that our model outperforms the SMM method in MHC-peptide binding prediction, with statistical significance. Our model also displays slightly better performance than the PickPocket method in leave-one-allele-out cross-validation. Experiment optimization is evaluated by restricting peptide selection to sets with data held out during model training, and assessing improvement in both model parameters and prediction performance on additional peptides with held-out data. Results show that our optimization strategy leads to selection of peptides that are statistically better than a random approach in improving MHC-peptide binding prediction.
Description	1 online resource (vi, 49 leaves) : illustrations (mostly color)
Format	application/pdf
College	Dartmouth College -- Department of Computer Science.
Issuing Body	Dartmouth Dissertations are made available through the Dartmouth Digital Library.
Degree	Thesis (Ph.D)--Dartmouth College.
Copyright	Made available by the author under a Creative Commons Attribution-NonCommercial License.; Made available by the Trustees of Dartmouth College under a Creative Commons Attribution-NonCommercial License. https://creativecommons.org/licenses/by-nc/4.0/
DOI	https://doi.org/10.1349/ddlp.2159

Find results with:

Searching collections:

Home Dartmouth Dissertations jiang2016 1

Improving MHC-peptide binding prediction from sparse data with a multi-allele graphical model and active learning / by Li Jiang

Object Description

Description

Improving MHC-peptide binding prediction from sparse data...

Improving MHC-peptide binding prediction from sparse data...

	All
	Pages Enter page numbers or page ranges separated by commas. For example: 1,3,5-12