The pipeline starts with data collection. RNA data is collected using mRNA-seq . Next the data goes through several preprocessing steps. First of which is aligning the data to the predicted gene model. Once the mapping is done, Map files are used to create count files, which are formatted to be used in weka. All of the data mining is done using Weka. The two mining approaches used are clustering and classification. From the mining techinques we are able to infer some useful meaning from the data.

The process of mRNA-Seq start with extracting RNA from the specimen of interest. Once the RNA is extracted a process known as reverse transcriptase is used to convert the RNA into cDNA. This is done because cDNA is much more stable and easier to work with. After being converted into cDNA the cDNA is sequenced using an Illumina mechine. More infromation on mRNA-Seq can be found here.

DATASET:
79 lanes (tissue samples)
~25,000 genes
~1 billion reads (50-150 base pairs)
5 different organs

Before meaning can be extracted from the reads, they must first be aligned to a reference genome or gene model. This enable the reads to be order in some meaningful way and also provides useful information about the gene expression level. RNA with higher expression level will have more sequence reads.

The count files are an account for how many reads for a particular gene are mapped to a specific tissues sample. Each lane or tissue sample can come from the same or a different tissue. For example, l_1 and l_2 could both come from liver, while l_3 could possibly be samples from kidney. So, there are multiple sample of the same tissue throughout the data set. The numbers in the columns represent the number of reads mapped to a particular tissue.

Once the count files are created they must be transposed and reformatted for Weka.