Data Mining

Data mining has been defined as the non-trivial extraction of implicit, previously unknown and potentially useful information from data. It should not be confused with bioinformatics which is more focused on sequence based extractions of specific patterns or motifs and also on specific pattern matching. Data mining compliments and expands bioinformatics, so that bioinformatics and data mining, are distinct although eventually they may merge together. (Eventually data mining and bioinformatics will be indistinguishable, but for the time being they are distinct).

Since lot of data is generated in biopharmaceutical industry, decision about the targets and lead compounds to be developed further is a major challenge. Data mining helps to make sense of these complex data sets in an intuitive and efficient manner. Many companies provide data mining services for a variety of purpose. There are six major approaches of data mining applications: (i) Influence-based mining – search for cause and effect relationships between data sets – pharmacogenomics. (ii) Affinity based mining – datamining system identifies data points or sets that tend to be grouped together; the approach is useful, and is important to distinguish “accidental/incidental” motifs from those of biological significance. (iii) Time-day data mining – look for patterns that are combined or rejected, as data set increases in future. (iv) Trends-based data mining – changes are studied that occur in specific data sets over time (trends are examined). (v) Comparative data mining- data collected at different sites over different time periods are compared to detect dissimilarities. (vi) Predicitive data mining – it complements and expands traditional bioinformatics