Amalgamation of protein sequence, structure and textual information for improving protein-protein interaction identification

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6396–6407 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 6396 Amalgamation of protein sequence, structure and textual information for improving protein-protein interaction identification Pratik Dutta, Sriparna Saha Department of Computer Science & Engineering Indian Institute of Technology Patna (pratik.pcs16, sriparna)@iitp.ac.in Abstract An in-depth exploration of protein-protein in- teractions (PPI) is essential to understand the metabolism in addition to the regulations of bi- ological entities like proteins, carbohydrates, and many more. Most of the recent PPI tasks in BioNLP domain have been carried out solely using textual data. In this paper, we argue that incorporation of multimodal cues can improve the automatic identification of PPI. As a first step towards enabling the de- velopment of multimodal approaches for PPI identification, we have developed two multi- modal datasets which are extensions and multi- modal versions of two popular benchmark PPI corpora (BioInfer and HRPD50). Besides, existing textual modalities, two new modali- ties, 3D protein structure and underlying ge- nomic sequence, are also added to each in- stance. Further, a novel deep multi-modal ar- chitecture is also implemented to efficiently predict the protein interactions from the devel- oped datasets. A detailed experimental anal- ysis reveals the superiority of the multi-modal approach in comparison to the strong baselines including uni-modal approaches and state-of the-art methods over both the generated multi- modal datasets. The developed multi-modal datasets are available for use at https:// github.com/sduttap16/MM_PPI_NLP. 1 Introduction Understanding protein-protein interactions (PPI) is indispensable to comprehend different biologi- cal processes such as translation, protein functions (Kulmanov et al., 2017), gene functions (Dutta and Saha, 2017; Dutta et al., 2019b), metabolic pathways, etc. The PPI information helps re- searchers to discover disease mechanisms and plays seminal role in designing the therapeutic drugs (Goncearenco et al., 2017). Over the years, a sig- nificant amount of protein-protein interaction in- formation has been published in scientific articles in unstructured text formats. However, in recent years, there has been an exponential rise in the number of biomedical publications (Khare et al., 2014). Therefore, it becomes imperative, urgent and of extreme interest to develop an intelligent information extraction system to assist biologists in curating and maintaining PPI databases. This pressing need has motivated Biomedi- cal Natural Language Processing (BioNLP) re- searchers to automatically extract PPI informa- tion by exploring various AI techniques. Re- cent advancements in deep learning (LeCun et al., 2015)(Bengio et al., 2007) have opened up new avenues in solving different well-known problems ranging from computational biology (Alipanahi et al., 2015; Dutta et al., 2019a), machine transla- tions (Cho et al., 2014), image captioning (Chen et al., 2017). Subsequently, there is a notable trend in using deep learning for solving different natural language processing (NLP) tasks in the biomedi- cal and clinical domains (Asada et al., 2018; Al- imova and Tutubalina, 2019) including the identifi- cation of protein-protein interactions from biomedi- cal corpora (Yadav et al., 2019; Peng and Lu, 2017). Multi-modal deep learning models, combining …

Content truncated to 500 words