Font Size: a A A

Protein Transduction Domain (ptd) Forecast Its Wear A Preliminary Study Of The Mechanism Of Membrane Function

Posted on:2006-11-24Degree:DoctorType:Dissertation
Country:ChinaCandidate:X L YangFull Text:PDF
GTID:1110360185473305Subject:Biomedical engineering
Abstract/Summary:PDF Full Text Request
Protein transduction domain (PTD) is a short peptide, which can ferry molecules across the cell membrane. Because of its potential to be a novel tool in gene and molecular therapies, PTD attract more interests in the past few years. However, the number of PTDs is still limited and the mechanism of penetrating the bio-membrane is not very clear by now.In this research work, SVM and TSVM algorithms were introduced to predict new PTDs from the peptide segments, which were extracted from SwissProt. These two algorithms' performances and the properties in the application were compared. And 1211 peptides were identified as new PTDs. Then, the conformational and physicochemical properties were analyzed on one of the predictions, hPer1-PTD. The related factors, which can affect the transductional function of hPer1-PTD, were found out.The main research works and results as follows:1. A bioinformatics software platform was established. This platform integrated not only several advanced bioinformatics analysis programs and essential databases, but also the analysis tools used in the following research. This platform provided the bioinformatics calculation to our lab.2. SVM and TSVM were used to predict PTDs.(1) A database of PTDs was built. This database recorded the data about PTDs, which have been proved by molecular biological experiments, and the potential PTDs, which were extracted from SwissProt.(2) A sequence encoding method, which was contrived according the characteristic of PTDs was established. Every PTD peptide segment was represented by 68 digital values. These values reflected the physicochemical and structural properties and their distribution in the PTDs' sequences.
Keywords/Search Tags:protein transduction domain, human period 1, machine learning, support vector machine, transductive support vector machine
PDF Full Text Request
Related items