AB-Amy: machine learning aided amyloidogenic risk prediction of therapeutic antibody light chains.

Over 120 FDA-approved antibody-based therapeutics are used to treat a variety of diseases.However, many candidates could fail because of unfavorable physicochemical properties. Light-chain amyloidosis is one form of aggregation that can lead to severe safety risks in clinical development. Therefore, screening candidates with a less amyloidosis risk at the early stage can not only save the time and cost of antibody development but also improve the safety of antibody drugs. In this study, based on the dipeptide composition of 742 amyloidogenic and 712 non-amyloidogenic antibody light chains, a support vector machine-based model, AB-Amy, was trained to predict the light-chain amyloidogenic risk. The AUC of AB-Amy reaches 0.9651. The excellent performance of AB-Amy indicates that it can be a useful tool for the in silico evaluation of the light-chain amyloidogenic risk to ensure the safety of antibody therapeutics under clinical development. A web server is freely available at http://i.uestc.edu.cn/AB-Amy/.

DeepACTION: A deep learning-based method for predicting novel drug-target interactions.

Drug-target interactions (DTIs) play a key role in drug development and discovery processes. Wet lab prediction of DTIs is time-consuming, expensive, and tedious. Fortunately, computational approaches can identify new interactions (drug-target pairs) and accelerate the process of drug repurposing. However, a vast number of interactions remain undiscovered; therefore, we proposed a deep learning-based method (deepACTION) for predicting potential or unknown DTIs. Here, each drug chemical structure and protein sequence are transformed according to structural and sequence information using different descriptors to represent their features correctly. There have been some challenges, such as the high dimensionality and class imbalance of data during the prediction process. To address these problems, we developed the MMIB technique to balance the majority and minority instances in the dataset and utilized a LASSO model to handle the high dimensionality of the data. In addition, we trained the convolutional neural network algorithm with balanced and reduced features for accurate prediction of DTIs. In this study, the AUC is considered a primary evaluation metric for comparing the performance of the deep ACTION model with that of existing methods by a 5-fold cross-validation test. Our experiential dataset obtained from the DrugBank database and our deepACTION model achieved an AUC of 0.9836 for this dataset. The experimental results ensured that the model can predict significant numbers of new DTIs and provide complete information to motivate scientists to develop drugs.

SSH: A Tool for Predicting Hydrophobic Interaction of Monoclonal Antibodies Using Sequences.

Therapeutic antibodies are one of the most important parts of the pharmaceutical industry. They are widely used in treating various diseases such as autoimmune diseases, cancer, inflammation, and infectious diseases. Their development process however is often brought to a standstill or takes a longer time and is then more expensive due to their hydrophobicity problems. Hydrophobic interactions can cause problems on half-life, drug administration, and immunogenicity at all stages of antibody drug development. Some of the most widely accepted and used technologies for determining the hydrophobic interactions of antibodies include standup monolayer adsorption chromatography (SMAC), salt-gradient affinity-capture self-interaction nanoparticle spectroscopy (SGAC-SINS), and hydrophobic interaction chromatography (HIC). However, to measure SMAC, SGAC-SINS, and HIC for hundreds of antibody drug candidates is time-consuming and costly. To save time and money, a predictor called SSH is developed. Based on the antibody's sequence only, it can predict the hydrophobic interactions of monoclonal antibodies (mAbs). Using the leave-one-out crossvalidation, SSH achieved 91.226% accuracy, 96.396% sensitivity or recall, 84.196% specificity, 87.754% precision, 0.828 Mathew correlation coefficient (MCC), 0.919 f-score, and 0.961 area under the receiver operating characteristic (ROC) curve (AUC).

CISI: A Tool for Predicting Cross-interaction or Self-interaction of Monoclonal Antibodies Using Sequences.

Monoclonal antibodies (mAbs) are one of the robust classes of therapeutic proteins. Their stability, specificity, and high solubility allow the successful development and commercialization of antibody-based drugs. Though with these characteristics, mAbs projects are often suspended due to self- or cross-interaction of monoclonal antibodies. This is one of the main reasons which causes the development of mAbs into drugs taking forever and expensive. CISI is short for cross-interaction or self-interaction of mAbs. It can be quantified by several assays. The assays such as poly-specificity reagent and cross-interaction chromatography can measure cross-interaction of mAbs. Self-interaction can be assayed through clone self-interaction by biolayer interferometry and affinity-capture self-interaction nanoparticle spectroscopy. To save time and money, we developed a model called CISI which can predict cross-interaction or self-interaction based on tripeptide composition. It showed 88.20% accuracy, 90.22% sensitivity, 86.05% specificity, 0.78 Mathew correlation coefficient, and 0.96 area under the receiver operating characteristic (ROC) curve (AUC) in the leave-one-out cross-validation. CISI is freely available at http://i.uestc.edu.cn/eli/cgi-bin/cisi.pl.

Development and Application of Computational Methods in Phage Display Technology.

BACKGROUND: Phage display is a powerful and versatile technology for the identification of peptide ligands binding to multiple targets, which has been successfully employed in various fields, such as diagnostics and therapeutics, drug-delivery and material science. The integration of next generation sequencing technology with phage display makes this methodology more productive. With the widespread use of this technique and the fast accumulation of phage display data, databases for these data and computational methods have become an indispensable part in this community. This review aims to summarize and discuss recent progress in the development and application of computational methods in the field of phage display. METHODS: We undertook a comprehensive search of bioinformatics resources and computational methods for phage display data via Google Scholar and PubMed. The methods and tools were further divided into different categories according to their uses. RESULTS: We described seven special or relevant databases for phage display data, which provided an evidence-based source for phage display researchers to clean their biopanning results. These databases can identify and report possible target-unrelated peptides (TUPs), thereby excluding false-positive data from peptides obtained from phage display screening experiments. More than 20 computational methods for analyzing biopanning data were also reviewed. These methods were classified into computational methods for reporting TUPs, for predicting epitopes and for analyzing next generation phage display data. CONCLUSION: The current bioinformatics archives, methods and tools reviewed here have benefitted the biopanning community. To develop better or new computational tools, some promising directions are also discussed.