A Novel Soft Clustering Approach for Gene Expression Data

Gene expression data represents a condition matrix where each row the gene and column shows condition. Micro array used to detect in lab for thousands of at time. Genes encode proteins which turn will dictate cell function. The production messenger RNA along with processing same are two main stages involved process expression. biological networks complexity added volume containing imprecision outliers increases challenges dealing them. Clustering methods hence essential identify patterns present massive data. Many techniques involve hierarchical, partitioning, grid based, density model based soft clustering approaches Understanding regulation other useful information from this can be possible only through effective algorithms. Though many discussed literature, we concentrate on providing approach analyzing population elements grouped fuzziness principle degree membership is assigned all elements. An improved Fuzzy by Local Approximation Memberships (FLAME) proposed work overcomes limitations while non-linear relationships provide better segregation functions.