CRF-LSTM Text Mining Method Unveiling the Pharmacological Mechanism of Off-target Side Effect of Anti-Multiple Myeloma Drugs

Abstract

Off-target effects played a vital role in the pharmacological understanding of drug efficacy and this research aimed to use text mining strategy to curate molecular level information and unveil the mechanism of off-target effect caused by the usage of anti-multiple myeloma (MM) drugs. After training a hybrid CNN-CRF-LSTM neural network upon the training data from TAC 2017 benchmark database, we extracted all of the side effects of 16 anti-MM drugs from drug labels, and combined the results with existed database. Afterwards, gene targets of anti-MM drugs were obtained by using structure similarity, and their related phenotypes were retrieved from Human Phenotype Ontology. Furthermore, linked phenotypes to candidate genes and adverse reaction of known drugs formed a knowledge graph. Through regulation analysis upon intersected phenotypes of drugs and target genes, an off-target effect caused by SLC7A7 was found, which with high possibility unveiled the pharmacological mechanism of side effect after using combination of anti-MM drugs.

Hybrid phenotype mining method for investigating off-target protein and underlying side effects of anti-tumor immunotherapy

“…Results for off-target proteins screening i) Side effect phenotype cross matching results. We use embedding similarity strategy [19] to match intersected terms in adverse reaction terms from drug labels and HPO terms from candidate off target proteins. As can be seen from Fig.…”

“…Furthermore, using the adverse effects of drug targets to establish a large-scale drug-target-adverse reaction network is a novel way to explain the ADR coincidence between target genes and drug target genes [18,19]. Owing to the rapid growth of biomedical literature in recent years, it is feasible to automatically extract knowledge from the published papers, such as the drugs and diseases they mention and their relations in a large scale [20,21].…”

“…In this paper, a pharmacological knowledge mining strategy iss proposed in the form of hybrid phenotypic mining by combining Biomedical Natural Language Processing (BioNLP) with medical informatics. First, we targeted at five PD-1/PD-L1 inhibitor drugs, and extract all the adverse drug reaction (ADR) from DailyMed drug labes by using conditional random fields (CRF) [19]. Second, we targeted at all the targeted proteins of the drugs, and extracted all the phenotype terms of the proteins from the Human Phenotype Ontology (HPO).…”

Adverse drug event detection using reason assignments in FDA drug labels

“…The supervised machine learning approaches utilize either Conditional Random Fields (CRFs) or Neural Networks (Deep Learning) approaches. For example, Zhou, et al 18 used a Convolutional Neural Network combined with Long Short Term Memory units with a CRF to identify side effects of 16 anti-Multiple Myeloma (MM) drugs from drug labels. The AutoMCExtractor 19 system used a set of CRF classifiers, trained on token, linguistic, and semantic features identified by cTAKES with a dictionary-based post-processing corrected boundary-detection errors of the CRF step.…”