Mitochondrial Protein Sequence-Computer Sciences Applications and Systems-Project Presentation, Slides of Applications of Computer Sciences

Download Mitochondrial Protein Sequence-Computer Sciences Applications and Systems-Project Presentation and more Slides Applications of Computer Sciences in PDF only on Docsity!

Outline of Presentation

 Introduction

 Basic Concepts

 Feature Extraction strategies

 Feature Selection strategies

 Classification techniques

 Description of Author’s results

 Comparison with Author’s results

 Comparison with other methods

 Conclusions

 Future Work

Basic Concepts

 Before starting discussion to our project one must have some basic concepts discussed as :

• Proteins
• Amino acids
• Mitochondrial protein sequence

Proteins

 Proteins are the major components of living organisms and constitute more than 25% weight of a cell.

 It performs functions e.g. catalysis, transport, transportation, digestion, movement, sensory capabilities, sense of taste, sense of vision and control the gene function.

 Proteins are made up of strings of amino acids(usually represented by English letters).

Amino acids

Amino acid 3-letter Abbreviation 1-letter Abbreviation Alanine Ala A Cysteine Cys C Aspartic acid Asp D Glutamic acid Glu E phenylalanine Phe F Glycine Gly G Histidine His H Isoleucine Ile I Lysine Lys K Leucine Leu L Methionine Met M Asparagine Asn N Proline Pro P Glutamine Gln Q Arganine Arg R Serine Ser S Threonine Thr T Valine Val V Tryptophan Trp W Tyrosine Tyr Y

Mitochondrial Protein Sequence

 >sp|P31937|3HIDH_HUMAN 3-hydroxyisobutyrate dehydrogenase, mitochondrial precursor (EC 1.1.1.31) (HIBADH) - Homo sapiens (Human).

MAASLRLLGAASGLRYWSRRLRPAAGSFAAVCSRSVASKTP

VGFIGLGNMGNPMAKNLMKHGYPLIIYDVFPDACKEFQDAGE

QVVSSPADVAEKADRIITMLPTSINAIEAYSGANGILKKVKKGS

LLIDSSTIDPAVSKELAKEVEKMGAVFMDAPVSGGVGAARSG

NLTFMVGGVEDEFAAAQELLGCMGSNVVYCGAVGTGQAAKI

CNNMLLAISMIGTAEAMNLGIRLGLDPKLLAKILNMSSGRCWS

SDTYNPVPGVMDGVPSANNYQGGFGTTLMAKDLGLAQDSA

TSTKSPILLGSLAHQIYRMMCAKGYSKKDFSSVFQFLREEETF

Data set

 The dataset used in this project has been generated in Jiang et al., 2006 and is received on request from [email protected].

 Comprises of 499 and 681 mitochondrial (positive) and non- mitochondrial (negative) sequences.

Feature Extraction Strategies

 We have used four kinds of proteins representations including:

• Amino acid composition (AAC).
• Pseudo amino acid composition (PseAAC).
• Dipeptide composition (Dp).
• Split amino acid composition (SAAC).

Pseudo amino acid composition

 Contains a set of greater than 20 discrete factors, where the first 20 represent the components of its conventional Amino Acid composition while the additional factors incorporate some sequence-order information via various modes. P = [ P 1 , P 2 ,…P 20 …. P20+ λ ]

 Whereas P 1 , P 2 ,…P 20 are the normalized occurrence frequencies of 20 amino acids and

 P 21 , P 22 , ……, PΛ are the 1st-teir to λ - tier correlation factor of amino acid sequence in the protein chain determined based on hydrophobicity and hydrophilicity.

 Hydrophobicity and hydrophilicity correlation functions are used

Dipeptide composition

 It is used to transform the variable length of proteins sequences to fixed length feature vectors.

 Occurrence frequency of every consecutive pair of amino acids is calculated F(i) = P(i) / P Where P(i) is the occurrence frequency of pair ‘ i’ and P is the total number of pairs in a protein sequence

 400 dimensional feature vector.

Split amino acid composition (Contd.)

 We have some very small number of sequences in the dataset where total amino acids in the protein sequence is less than 50.

 So for that we divided it into the same three parts but with:

• 10 amino acids of N termini
• 10 amino acids C termini and
• the region between these two terminuses.

Feature Selection Strategies

 We have employed two different feature selection strategies for dipeptide composition including:

• Rank Features Selection
• Features Selection Through GA

Testing Methods

 Jackknifing test

• One of the protein sequence pattern is taken as test sample and remaining N-1 sequence patterns are considered as training patterns.
• Label of a test sample is predicted using the rest of the N- training patterns.
• This process is repeated for all N patterns.

 Independent test

 Self Consistency test

Performance Measurements

 Sensitivity(), Specificity(), Accuracy (ACC), Mathew Correlation Coefficient (MCC)

• Sensitivity = TP/ (TP + FN)
• Specificity = TP/ (TP + FP)
• Acc = (TP + TN) / (TP + TN + FP +FN)
• MCC = TP.TN-FN.FP/Sqrt {(TP+FN)(TP+FP)(TN+FN)(TN+FP)} TP = true positive TN = true negative FP = false Positive FN = false negative