Article Preview
TopIntroduction
Nucleic acid ordering followed in polynucleotide chains contain vital information required to learn about a person’s hereditary properties. Measuring such sequences will provide valuable insights on identifying anomalies, hence will help in the process of curing disease, the process of identifying genome sequences for specific organisms is called DNA sequencing. Several fields utilizing DNA sequencing includes; disease diagnostics, biotechnology, forensic biology and biological systematics. The introduction of DNA sequencing techniques has been instrumental in accelerating biological research and discovery. Faster sequencing has enabled the construction of human genome in the Human Genome Project. Several related projects have alsobeen funded and have been instrumental in sequencing genomes of other comparatively smaller species such as plants, animals and microbes.
However, problems still exist in identifying and matching genome sequences. The computational problems existing in sequencing genomes is challenging both due to the complexity of the sequencing process and the huge amount of information to be processed for sequencing. Several techniques of extracting DNA sequences from a DNA sample are available. Illumina (Voelkerding et al., 2009) is one of the most recent and the most powerful sequencers available now. However due to the hugeness of the data involved, even the most powerful sequence extractors are able to extract only a sub section of the DNA rather than the entire DNA. This brings in the problem of sequencing these reads to formulate the structure of the DNA. Several problems are associated with sequencing, such as; sequence reads have large number of overlaps with each other, the generated sequence reads are unordered and they might contain several missing bases and experimental errors. This makes the sequencing process a difficult task. This approach deals with sequencing a genome using a reference genome and the sequence reads. Using a reference genome provides an effective guideline for the sequencing process to flow. Since genomes corresponding to similar organisms are mostly identical, this process becomes possible. It has been identified that 98% of a genome corresponding to same organisms are identical and variations occurs in the remaining 2% of the genome. This property has both advantages and disadvantages. The advantages raise from the fact that having a reference genome will act as a necessary guideline during the sequence alignment process. The disadvantage is that the occurrence of variations is not defined. Mutations from the referencing genome can occur at any point in the genome to be sequenced. This leads to the requirement of an approximate matching algorithm rather than a perfect matching algorithm.
The presented approach proposes an approximate matching technique for genome alignment using sequence reads using Map Reduce paradigm on Hadoop. The proposed technique is reference based, hence uses a reference genome and performs the process of fragment assembly using subsequence-matching techniques. The level of matches are identified and the fragment exhibiting high overlapping levels is fixed in the corresponding location. However, many such fragments can be identified and they may not be similar. HenceGreedy approximate matching technique is used to construct the final genome.
The paper is structured as follows, section 2 presents the related works, analyzing the sequencing generations and the current sequencing and fragment assembly techniques, section 3 presents a detailed discussion on subsequence based fragment assembly in genome using approximate matching techniques, section 4 presents the results obtained by assembling fastq data of Echinopstelfairi, Biomphalariaglabrata and Sarcophilusharrisii. The measures used for analysis are also discussed in detail and section 5 concludes the study and proposes future directions for faster and qualitative fragment assembly.