An iterative approach of protein function prediction

Abstract

BackgroundCurrent approaches of predicting protein functions from a protein-protein interaction (PPI) dataset are based on an assumption that the available functions of the proteins (a.k.a. annotated proteins) will determine the functions of the proteins whose functions are unknown yet at the moment (a.k.a. un-annotated proteins). Therefore, the protein function prediction is a mono-directed and one-off procedure, i.e. from annotated proteins to un-annotated proteins. However, the interactions between proteins are mutual rather than static and mono-directed, although functions of some proteins are unknown for some reasons at present. That means when we use the similarity-based approach to predict functions of un-annotated proteins, the un-annotated proteins, once their functions are predicted, will affect the similarities between proteins, which in turn will affect the prediction results. In other words, the function prediction is a dynamic and mutual procedure. This dynamic feature of protein interactions, however, was not considered in the existing prediction algorithms.ResultsIn this paper, we propose a new prediction approach that predicts protein functions iteratively. This iterative approach incorporates the dynamic and mutual features of PPI interactions, as well as the local and global semantic influence of protein functions, into the prediction. To guarantee predicting functions iteratively, we propose a new protein similarity from protein functions. We adapt new evaluation metrics to evaluate the prediction quality of our algorithm and other similar algorithms. Experiments on real PPI datasets were conducted to evaluate the effectiveness of the proposed approach in predicting unknown protein functions.ConclusionsThe iterative approach is more likely to reflect the real biological nature between proteins when predicting functions. A proper definition of protein similarity from protein functions is the key to predicting functions iteratively. The evaluation results demonstrated that in most cases, the iterative approach outperformed non-iterative ones with higher prediction quality in terms of prediction precision, recall and F-value.

Protein function prediction using weak-label learning

“…al. [6] proposed an iterative protein function prediction method using partial annotations. At each iteration, using the most confident predicted functions, pairwise similarities between training proteins and testing proteins are updated.…”

Protein Function Prediction with Incomplete Annotations

“…This idea is also used in Chi et al [37] and Wang et al [7]. Thus we can use the function set of a protein to enrich its feature representation.…”

“…We compare our methods with (i) WELL [24], (ii) MLR-GL [25], (iii) FCML [7], and (iv) CIA [37]. The first two approaches are weak-label learning methods, and the other two methods are recently developed protein function prediction algorithms using multilabel learning and PPI networks.…”

“…There are many other ways to define the function correlations, e.g., Jaccard coefficient [16] and Lin's similarity [29]. Here, we use the cosine similarity for its simplicity and wide use [6], [7], [37]. Note that Eq.…”

“…Re et al [36] developed an efficient ranking based prediction model using local and global learning strategies. Chi et al [37] proposed an iterative protein function prediction method called Cosine Iterative Al-gorithm (CIA). CIA increases the number of predicted functions on unlabeled proteins iteratively.…”