Shayan Ahmed, Jamia Millia Islamia, New Delhi
Polypharmacy:
Polypharmacy refers to the usage of several medicines at the same time. It is the combination of drugs, and has become a frequent practice in modern medicine, particularly among the elderly and patients with complicated illnesses. When two or more drugs are taken at the same time, they may interact. Polypharmacy may be more successful in treating illnesses, but unanticipated drug-drug interactions (DDIs) might occur.
Drug-Drug interactions:
DDI is defined as a change in the pharmacologic action of one medication when combined with another. DDIs are the most prevalent cause for patients to visit emergency rooms and can result in Adverse Drug Reactions (ADRs) (i.e., side effects) including mortality, making it a major public health issue. Most documented polypharmacy adverse effects are uncommon, and they are seldom found in small clinical studies. As a result, manually identifying these side effects is challenging. As a result, developing computational approaches for forecasting DDIs is required.
Problems in DDI detection:
The approaches in the DDI prediction problem are classified into two types. The first category just detects the existence or absence of interactions and does not identify the kind of adverse effects. These techniques collect interactions through experiments and clinical research, medical records, and network modeling based on DDIs, side effects, and structural similarities. The objective of the second category, on the other hand, is to determine the sort of adverse effects that exist between medicines. The approaches in the second category play their role in reducing the impact of polypharmacy side effects. In a recent study, researchers developed a neural network-based method for polypharmacy side effects prediction (NNPS).
NPPS as a tool against DDI:
Understanding DDI side effects is a critical step in medication development and drug co-administration. As a result, several computer techniques for forecasting polypharmacy side effects are being developed. In the NNPS approach, each drug is represented by a feature vector based on mono side effects and drug-protein interactions, and the Principal Components Analysis (PCA) is used for dimension reduction of feature vectors to reduce method complexity.
The relevant drug feature vectors for a specific medication combination are added together to train the neural network for predicting polypharmacy side effects. NNPS outperforms the competition for two reasons. The new feature vectors created by dimension reduction methods are the first and most important factors. The second reason is that a straightforward neural network design was adopted.
Results & Significances:
NNPS is a quick and effective method for examining a large number of polypharmacy adverse effects. For 964 polypharmacy adverse effects, NNPS is compared to five well-known methods: Decagon, Concatenated drug characteristics, Deep Walk, DEDICOM, and RESCAL. The NNPS was compared to five well-known techniques in terms of accuracy, complexity, and running time speed, and the results demonstrate that the given method performs well for a critical and difficult topic in pharmacology.
Also read: An intense embrace of two giants in cell division
References:
- Masumshah, R., Aghdam, R., & Eslahchi, C. (2021). A neural network-based method for polypharmacy side effects prediction. BMC bioinformatics, 22(1), 385. https://doi.org/10.1186/s12859-021-04298-y
- The Corrosion Prediction from the Corrosion Product Performance
- Nitrogen Resilience in Waterlogged Soybean plants
- Cell Senescence in Type II Diabetes: Therapeutic Potential
- Transgene-Free Canker-Resistant Citrus sinensis with Cas12/RNP
- AI Literacy in Early Childhood Education: Challenges and Opportunities
Author info:
Shayan Ahmed is currently pursuing a Master of Science degree in Microbiology from the Department of Biosciences, Jamia Millia Islamia, New Delhi. His area of research interest lies in antibiotic resistance and associated molecular mechanisms. His recent work was focused on understanding colistin resistance patterns in the environment, particularly in water bodies.
Other Publications by Author:
Prognostic prediction score: A method to predict Pleuroparenchymal fibroelastosis
Soumya Shraddhya Paul, Amity University, Noida Pleuroparenchymal fibroelastosis (PPFE) is an uncommon form of interstitial lung disease (ILD) that is characterised by elastofibrosis in the upper lobes. Patients with PPFE have several distinct clinical characteristics not seen in other ILDs. These are gradual weight loss and restrictive ventilatory dysfunction with increasing residual volume (RV). PPEE […]