A joint optimization framework integrated with biological knowledge for clustering incomplete gene expression data

Clustering algorithms have been successfully applied to identify co-expressed gene groups from expression data. Missing values often occur in data, which presents a challenge for clustering. When partitioning incomplete data into groups, missing value imputation and clustering are generally performed as two separate processes. These two-stage methods likely result unsuitable task unsatisfying performance. This paper proposes multi-objective joint optimization framework that addresses this problem. The proposed can impute the under guidance of clustering, therefore realize synergistic improvement In addition, similarity semantic extracted Gene Ontology combined, form functional neighbor interval each value, provide reasonable constraints framework. experiments carried out on several benchmark sets. terms average rate over sets different rates, our reduce error by 6.4–14.7% increase accuracy 4.0–10.1% compared with six popular promising methods. Furthermore, biological significance identified clusters is reported evaluate effectiveness