Prediction of protein 3D structure from contact map

العناوين الأخرى

توقع الشكل الثلاثي للبروتين من خريطة الاتصال

مقدم أطروحة جامعية

al-Qallab, Nawal T.

مشرف أطروحة جامعية

al-Nihoud, Jihad Quball Awdah
al-Fawarih, Hamid J.

أعضاء اللجنة

al-Rababiah, Mamun S.
Nusayr, Maryam
Hamarishah, Akram

الجامعة

جامعة آل البيت

الكلية

كلية الأمير الحسين بن عبد الله لتكنولوجيا المعلومات

القسم الأكاديمي

قسم علوم الحاسوب

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2009

الملخص الإنجليزي

Bioinformatics is an interesting topic for both the biologists and computer scientists.

The ability to predict protein 3D structure from the amino acid sequence is not less than revolutionize in this area.

A fundamental principle in all protein sciences is that protein structure leads to protein function.

The 3D structure of protein can be represented using N*N symmetrical binary matrix C called contact map whose element C (i, j) =1 if and only if the physical distance between amino acid i and/ is less than or equal to a pre assigned threshold T otherwise C (i, j) = 0 Predicting 3D structure directly from primary sequence of protein is very complex problem, so various computational approach participate in analyzing and extracting rules to predict tertiary structure from contact map.

This thesis focuses on developing a method for predicting 3D structure of protein from contact map using MATLAB, which is contract with mathematical properties that can be derived from distance values between pairs of the amino acid, typically measured in Angstroms 04°).

The proposed method focuses on choosing the threshold value for computing the contact map, which is affecting connectivity between the contact map and its 3D structure, not any threshold give accurate contact map which provides exact 3D structure.

We found that the contact maps computed using threshold values (12-18) A allow better 3D structure recovery than those computed at thresholds (7-9) A.

This approach aims to detect the dense areas that form the basic functional areas in the contact map by scanning module.

Looking for the dense area is an important step that will improve the performance of the predicting 3D structure of protein from it CM based on prediction quality more than quantity of contacts.

The experimental results in this thesis are obtained on data set of proteins related to different classes (mainly alpha, mainly beta, mixed alpha and beta) which are extracted from PDB.

The average execution time of the proposed algorithm is varying depending on protein size.

The results show that the predicting structures can be determined using MATLAB in reliable and efficient manner.

التخصصات الرئيسية

الأحياء

الموضوعات

• علم الأحياء
• الهيكل الاجتماعي
• علم الأحياء الجزيئي
• البروتينات
• معالجة البيانات
• المعلوماتية الحيوية

عدد الصفحات

64

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-321576