Pf-Phospho: a machine learning-based phosphorylation sites prediction tool for Plasmodium proteins

Abstract

Even though several in silico tools are available for prediction of the phosphorylation sites for mammalian, yeast or plant proteins, currently no software is available for predicting phosphosites for Plasmodium proteins. However, the availability of significant amount of phospho-proteomics data during the last decade and advances in machine learning (ML) algorithms have opened up the opportunities for deciphering phosphorylation patterns of plasmodial system and developing ML-based phosphosite prediction tools for Plasmodium. We have developed Pf-Phospho, an ML-based method for prediction of phosphosites by training Random Forest classifiers using a large data set of 12 096 phosphosites of Plasmodium falciparum and Plasmodium bergei. Of the 12 096 known phosphosites, 75% of sites have been used for training/validation of the classifier, while remaining 25% have been used as completely unseen test data for blind testing. It is encouraging to note that Pf-Phospho can predict the kinase-independent phosphosites with 84% sensitivity, 75% specificity and 78% precision. In addition, it can also predict kinase-specific phosphosites for five plasmodial kinases—PfPKG, Plasmodium falciparum, PfPKA, PfPK7 and PbCDPK4 with high accuracy. Pf-Phospho (http://www.nii.ac.in/pfphospho.html) outperforms other widely used phosphosite prediction tools, which have been trained using mammalian phosphoproteome data. It also has been integrated with other widely used resources such as PlasmoDB, MPMP, Pfam and recently available ML-based predicted structures by AlphaFold2. Currently, Pf-phospho is the only bioinformatics resource available for ML-based prediction of phospho-signaling networks of Plasmodium and is a user-friendly platform for integrative analysis of phospho-signaling along with metabolic and protein–protein interaction networks.

Post-Translational Modifications of Proteins of Malaria Parasites during the Life Cycle

“…A machine learning approach was applied to predict new phosphorylation sites in proteins of P.f. [ 361 ], whose results are the theoretic base for further experimental studies to confirm, or not, the prediction. Note that the majority of studies were performed with the most dangerous human and the rodent malaria parasites, P.f.…”

Complementary crosstalk between palmitoylation and phosphorylation events in MTIP regulates its role during Plasmodium falciparum invasion

“…To understand the interdependence between PTMs and its effect on the parasite life cycle, we began by annotating a pool of proteins within the P. falciparum 3D7 proteome, which are susceptible to either phosphorylation or palmitoylation. A cumulative of 2,148 “high confidence” phosphoproteins could be compounded from curated sources ( Ganter et al., 2017 ; Gupta et al., 2022 ), systematic search results (search term = “phosphoprotein” in UniProtKB, NCBI, and PlasmoDB), and Pf-Phospho prediction results ( Supplementary Table 2 ). Phosphoproteins predicted by Pf-Phospho were labeled “Phosphorylable,” and those retrieved from curated resources and databases, using systematic searches, were labeled as “Phosphorylated.” In addition, 3,105 proteins were either predicted ( Ren et al., 2008 ) or found in curated datasets ( Jones et al., 2012 )/databases (NCBI, UniProtKB, and PlasmoDB) to be components of the parasite palmitome ( Supplementary Table 3 ).…”

“…The updated proteome of P. falciparum 3D7 was indexed from databases like PlasmoDB Release 58 ( Amos et al., 2022 ), National Center for Biotechnology Information or NCBI (GCA_000002765), Pf-Phospho ( Gupta et al., 2022 ), and UniProtKB Release 2022_02 ( Bateman et al., 2021 ), and also from other curated resources ( Solyakov et al, 2011 ; Jones et al, 2012 ; Lasonder et al, 2012 ; Govindasamy et al, 2016 ; Kumar et al, 2017a ; Kumar et al, 2017b ; Pease et al, 2018 ; Blomqvist et al, 2020 ; Gupta et al, 2022 ). Probable pseudogenes were removed from the library, and only the genes with a protein coding potential were selected for downstream processing.…”

“…Phosphorylation events specific to the proteome assembly were retrieved from curated mass spectrometry resources ( Ganter et al., 2017 ), databases (PlasmoDB, UniProtKB, PfPhospho, and NCBI), and Pf-Phospho ( Gupta et al., 2022 ) prediction results. Proteome-specific palmitoylation events and sites were cataloged from CSS-Palm prediction results ( Ren et al., 2008 ), publicly available mass spectrometry resources ( Jones et al., 2012 ), and systematic database search results from PlasmoDB, NCBI, and UniProtKB.…”

Machine learning for predicting Plasmodium liver stage development in vitro using microscopy imaging

“…Artificial intelligence using machine learning has been effectively used for various tasks in malaria research: prediction : predicting human-parasite protein associations using network topological profiles and machine learning [9] , parasite protein phosphosite prediction using machine-learning [10] ; detection : detecting malaria parasites using light microscopy and convolutional neural networks (CNNs) [11] , [12] , using patient information from parasite case reports, and machine learning techniques [13] , using patient symptoms and demographic features together with machine learning models [14] , using protein sequences and natural language processing techniques [15] , or using bead-based antigen detection assay in conjunction with decision trees [16] , and drug discovery [17] .…”