Signal-BNF: a Bayesian network fusing approach to predict signal peptides

In silico high throughput mutagenesis and screening of signal peptides to mitigate N-terminal heterogeneity of recombinant monoclonal antibodies

N-terminal heterogeneity resulting from non-uniform signal peptide (SP) cleavage can potentially affect biologics property attributes and result in extended product development timelines. Few studies are available on engineering SPs systematically to address miscleavage issues. Herein, we developed a novel high throughput computational pipeline capable of generating millions of SP mutant sequences that uses the SignalP 5.0 deep learning model to predict which of these mutants are likely to alleviate the N-terminal miscleavage in antibodies. We optimized the parameters to target mutating one or two amino acids at the C-terminus of 84 unique SPs, exhausting all theoretically possible combinations and resulting in a library of 296,077 unique wildtype and mutant signal peptides for in silico screening of each antibody. We applied this method to five antibodies against different targets, with various extent of miscleavage (2.3% to 100%) on their Lambda light chains. In each case, multiple SP mutants were generated, with miscleavage reduced to a non-detectable level and titers comparable with or better than that of the original SPs. Pairwise mutational analysis using an in silico library enriched with high-scoring mutants revealed patterns of amino acids at the C-terminus of SPs, providing insights beyond the “Heijne rule”. To our knowledge, no similar approach that combines high throughput in silico mutagenesis and screening with SP cleavage prediction has been reported in the literature. This method can be applied to both the light chain and heavy chain of antibodies, regardless of their initial extent of miscleavage, provides optimized solutions for individual cases, and facilitates the development of antibody therapeutics.

Abbreviations: Aa, amino acids; CHO, Chinese hamster ovary; CNN, convolutional neural network; CSscore, cleavage site score; CSV, comma-separated values; HC, heavy chain; HEK, human embryonic kidney; HPLC, high-performance liquid chromatography; IgG, immunoglobulin G; IGLV, immunoglobulin G Lambda variable; LC, light chain; LCMS, liquid chromatography–mass spectrometry; MS, mass spectrometry; PCR, polymerase chain reaction; PBS, phosphate-buffered saline; PEI, polyethylenimine; SP, signal peptide; SPase, signal peptidase; TCEP, tris(2-carboxyethyl) phosphine; TOF, time-of-flight.

Prediction of Signal Peptides in Proteins from Malaria Parasites

Signal peptides are N-terminal presequences responsible for targeting proteins to the endomembrane system, and subsequent subcellular or extracellular compartments, and consequently condition their proper function. The significance of signal peptides stimulates development of new computational methods for their detection. These methods employ learning systems trained on datasets comprising signal peptides from different types of proteins and taxonomic groups. As a result, the accuracy of predictions are high in the case of signal peptides that are well-represented in databases, but might be low in other, atypical cases. Such atypical signal peptides are present in proteins found in apicomplexan parasites, causative agents of malaria and toxoplasmosis. Apicomplexan proteins have a unique amino acid composition due to their AT-biased genomes. Therefore, we designed a new, more flexible and universal probabilistic model for recognition of atypical eukaryotic signal peptides. Our approach called signalHsmm includes knowledge about the structure of signal peptides and physicochemical properties of amino acids. It is able to recognize signal peptides from the malaria parasites and related species more accurately than popular programs. Moreover, it is still universal enough to provide prediction of other signal peptides on par with the best preforming predictors.

A survey of computational intelligence techniques in protein function prediction

During the past, there was a massive growth of knowledge of unknown proteins with the advancement of high throughput microarray technologies. Protein function prediction is the most challenging problem in bioinformatics. In the past, the homology based approaches were used to predict the protein function, but they failed when a new protein was different from the previous one. Therefore, to alleviate the problems associated with homology based traditional approaches, numerous computational intelligence techniques have been proposed in the recent past. This paper presents a state-of-the-art comprehensive review of various computational intelligence techniques for protein function predictions using sequence, structure, protein-protein interaction network, and gene expression data used in wide areas of applications such as prediction of DNA and RNA binding sites, subcellular localization, enzyme functions, signal peptides, catalytic residues, nuclear/G-protein coupled receptors, membrane proteins, and pathway analysis from gene expression datasets. This paper also summarizes the result obtained by many researchers to solve these problems by using computational intelligence techniques with appropriate datasets to improve the prediction performance. The summary shows that ensemble classifiers and integration of multiple heterogeneous data are useful for protein function prediction.