Muneer Ahmad

An Integrated Approach for Protein Structure Prediction Using Artificial Neural Network

Protein prediction is a fundamental problem in Bioinformatics. Protein structure prediction has vital importance in drug design and biotechnology. Huge amount of biological importance data is being produced and there is great need to transcribe the DNA sequences into amino acid sequences because peptide functions perform important role in body functions of species. Exponential growth of genomic data and complex structure of protein make it challenging to predict its structure. In this paper, we are proposing an integrated approach for the prediction of tri-nucleotide base patterns in DNA strands leading to transcription of peptide regions in genomic sequences. The approach comprise of preprocessing of data, transcription engine and post processing of output. The task has been carried out using series of filters that purify the raw data and assign weights to bases for further feeding to central engine. JOONE (Java Object Oriented Neural network) takes input in the form of segmented data and assign to processes at sigmoid layers. Each layer contains processes and feed forward and back propagation techniques make it possible to predict the sample pattern from genomic sequences of variant sizes.

A comprehensive survey on genome sequence analysis

This paper is an attempt to highlight the fundamental approaches, algorithms and techniques being employed in interesting areas of Computational Biology. The comparative analysis of genomic sequences, micro-array technology, phylo-genetic tree, alignment matters, maximum sequence search and protein structure prediction are some of NP hard problems. There has been no direct solution proposed so far. Approximate and optimal solutions have been devised using different tools. We have compared different approaches in various domains with the aid of exemplary data. We assigned different weight values for relevant features and obtained scores in comparison and alignment phenomenon. We classified sequence analysis to be a broader domain encapsulating visualization, pattern recognition, motif concentration and structure prediction. Different techniques and approaches have been evaluated and compared. It was analyzed that some approaches suitable for certain domains may not be recommended to employ for other domains. The usefulness of this analysis has been explored and further future expectations have been described.

An integrated statistical comparative analysis between variant genetic datasets of Mus musculus

Comparative genomic analysis between variant datasets of same specie is considered to be vital to discover the degree of relevancy in them. This analysis helps in the categorisation of diversity of features in species. An immense need was felt to build sophisticated tools for efficient and robust comparative analysis. The accuracy of methodologies is directly proportional to sensitivity involved in comparing datasets for optimality. This paper is a depiction of an effort for the discovery of variant features between genetic datasets of Mus musculus. The approach described is demonstrated phase-wise with the inclusion of specific filters at each stage. At first instance, cleansing filter refines the datasets. Further series of filters depict the layered process for comprehensive comparative analysis. Numerical results have been evaluated. The protein translation phase has been introduced with conceptual demonstration of codon composition phenomenon. Characteristics of density, nucleotide strengths and codon composition better reflect the relevancy in genetic datasets of Mus musculus.

Breast Carcinoma Pigeonholing and Vaticination Using an Interspersed and Malleable Approach

Breast carcinoma is considered as the second major cause of death in females. Malignant tumor affects some tissues of breast and may spread over neighboring tissues. Early detection of this malignant mass is very important to save the precious lives. Although the death rate is reduced by application of modern tools yet research for optimal solutions is still in progress to bring more comprehensive mechanisms. In this paper, we are proposing an interspersed approach for breast tumor pigeonholing and vatic nation. We trained our neural network over datasets obtained from the University of Wisconsin Hospitals, Madison and tested over many other datasets with diverse network architectures. The proposed approach was sectioned in applications of data filters. Our network architecture showed 96% of malignant and 99.45% of benign diagnosis for training confusion matrix and 100% for malignant and 97% benign for cross validation matrix. We have given detailed experimentations in light of training and cross validation mean square errors and demonstrated results even for minute curve fluctuations.

A better way for exon identification in DNA splicing

The strands of chromosomes are supposed to be split into genic and intergenic regions. Gene identification is an optimization problem in Eukaryotic DNA splicing. An optimal solution is essential which may help in better protein translation. We have proposed a novel approach for gene identification by employing discrete wavelet transforms for noise reduction in DNA sequences and introducing a new indicator sequence for better signal generation. Wavelet transforms greatly reduced the background noise and visible peaks of genic regions were found in power spectral estimation. The comparative analysis of proposed and existing approaches showed significant results for novel approach over prevailing solutions over dataset Yersinia pestis (ACCESSION: NC_004088, 4000 bp, four genes and exons from location 5000 bp to 8999 bp). Similar outperformace was observed over many other datasets.

Lung carcinoma pigeonholing and vaticination by interspersed approach

Vaticination and pigeonholing of lung carcinoma is a conspicuous nuisance for emancipate genesis. The carnage rate in elderly age is high as compared to younger ones and cure from carcinoma at premier stage is most salubrious. In this paper, we are proposing an interspersed and malleable approach for lung carcinoma pigeonholing and vaticination. The contrivance has been crumbled into laconic precincts that assimilate the visionary affirmation of pigeonholing and vaticination prodigy over training and testing datasets. This conviction serves as substratum for multi dimensional view / pigeonholing of aberrant datasets. We obtained 100 percent results for training confusion matrix and 83 percent for cross validation confusion matrix. The peerless least mean square error remained below 0.05 which shows a well trained architecture over multiple data segments.

A Pattern Matching Technique for Multiple Sequences Alignment with GAP Consideration

Protein and DNA sequences of different organisms are often related and they indicate the knowledge about species. The consideration is made to align more than two sequences so that a level/extend of similarity or differences be found that would help in categorizing the common characteristics of species and their behaviors. An efficient recursive approach is proposed in this paper that would not only find the multiple sequences Alignment for protein/DNA sequence but also provides means for consideration of gaps between them. The algorithm will calculate the degree of similarity and bounds/extends of gaps to bring refined and useful results. The input variables (e.g. Strands) of the program are user dependant and internal calculations are performed in recursive fashion to add intelligence to the input Strands. Experimental results have shown more favorable performance of the proposed approach as compared to other approaches.