**1. Introduction**

The field of research involving analysis of the functioning of drugs in the human body is referred to as pharmacology. It details a class of medical science intended to address the employment of drugs in preventing, analysing, and treating diseases. As such, pharmacologists research the chemical properties of drugs and their impacts on the human body, in order to develop novel drugs and test them for efficacy and safety. Toxicology is the study of poisons, and relevant research helps to identify the effects of toxins and toxicants on the human body. It is a branch of pharmacology that aids in addressing the characterisation, identification, and quantification of the biological agents and chemicals that are responsible for creating adverse effects of the chemical agents on the human body. They develop novel methods to test for toxicity, and identify and quantify the toxicants in provided samples [1,2]

**Citation:** Latha Bhaskaran, K.; Osei, R.S.; Kotei, E.; Agbezuge, E.Y.; Ankora, C.; Ganaa, E.D. A Survey on Big Data in Pharmacology, Toxicology and Pharmaceutics. *Big Data Cogn. Comput.* **2022**, *6*, 161. https://doi.org/10.3390/ bdcc6040161

Academic Editors: Domenico Talia and Fabrizio Marozzo

Received: 13 October 2022 Accepted: 2 December 2022 Published: 19 December 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Pharmaceutics is a branch of pharmacy that manages the process of turning new chemical agents or old drugs into medications that can be used safely and efficiently by patients. It is also referred to as the science of dosage, which informs the design. Many chemicals with pharmacological properties require special measures to assist in attaining therapeutically related amounts at their sites of action. Pharmaceutics also helps in relating the formulations of drugs to their disposition and delivery in the body. Pharmacology, pharmaceutics, and toxicology are all categorised as interdisciplinary sciences [3,4]. They involve the extraction of knowledge from several disciplines, such as chemistry, biology, mathematics, and physics. BD has been one of the most researched topics in recent decades. BD denotes large or complex data sets that render the use of existing data-processing methods inadequate. Challenges associated with BD analysis include data storage, data capturing, sharing, searching, visualising, transferring, information privacy, and data source updating.

BD in health and medical care has attracted the attention of researchers owing to its various benefits. BD-based solutions for clinical decision support systems have revealed promising results in treating diseases such as Alzheimer's disease [5]. BD approaches have also been used for optimising electronic medical records [6]. The data generated in pharmacology, toxicology, and pharmaceutics include outcomes regarding the clinical behaviours, pharmacokinetics, chemical properties, associated toxicity of drugs. Data from cell culture experiments and animal studies are also included in this category. Thus, the volume of data generated in pharmaceutics, pharmacology, and toxicology has been increasing rapidly. This is primarily due to the employment of next-generation sequencing (NGS) and high-throughput screening (HTS). HTS is a technology that enables the simultaneous testing of millions of chemical elements. Additionally, NGS enables the efficient sequencing of RNA or DNA molecules.

It has been estimated that the global data volume generated by the toxicology, pharmaceutics, and pharmacology fields is approximately 2.5 zettabytes, where one zettabyte is equal to one sextillion bytes. BD is generally useful in pharmacology, pharmaceutics, and toxicology, as it can help to improve the prediction rate and accuracy of the effects of the researched drugs. It has helped to enhance the accuracy and speed of drug discovery and development. Although there are certain associated advantages of using BD in pharmacology, toxicology, and pharmaceutics, there are also certain drawbacks. BD can be challenging and overwhelming to navigate, and it may be difficult to trace useful data from the vast database. In addition, the implementation of BD is generally expensive, due to the overheads related to data collation, transformation, and analyses/modelling. The challenges and the overwhelming nature of BD can be resolved through the use of a data visualisation tool. Such an approach allows users to see the data and filter the necessary data, making it easier to understand, navigate, and extract the specific necessary information. Figure 1 shows a generic workflow representation of BD in the Hospital and Pharmacology ecosystem, based on the studies [7–9].

Hospital and clinics are the touchpoints for patients, that provide the symptom analysis, pre-diagnosis, physician advice, case history, and preliminary investigations from both the patient and physician perspectives. The data repository is the data store that collates data from all entities in the ecosystem, which are stored for further analytics, and as data sets used for prediction. In the scenarios presented in Figure 1, hospitals, clinics, and pharmacies provide the data that will eventually be stored in the big data repository. The repository possesses a big data nature, due to the data velocity and the heterogeneity of the data that are collated. The data from the BD repository are used for analytics such as identifying illness patterns, trends, and pre-emptive prediction of pandemics. Furthermore, BD analytics can be used to develop predictive models, which may be used to identify high -risk patients. Predictive model and pattern analytics can be used to model and develop targeted interventions. The use of a monitoring system coupled with BD models can form a proactive monitoring platform that can provide feedback to physicians. Based

on Figures 1 and 2, a graph-based representation of Scenario 1—Drug efficacy monitoring system using BD—was derived.

**Figure 2.** Flowchart for drug efficacy monitoring system using big data.

Scenario 1 considers a dataflow graph path for the drug efficacy monitoring system, one of the applications of BD in the healthcare domain. 1. The patient is attended at the clinic for pre-investigation; 2. the patients referred for specialist consultation are received at the hospital; 3. the required prescribed drug is collected from the pharmacy; 4. data of all patient activities are collected in the BD repository; 4.1. data cleaning and transformation are carried out as a continuous process; 6. the trend of the patient is analysed; 9. the patient is monitored in real-time during the drug course; 14. check whether hospitalisation

is required; 15. drug efficacy is improved based on all the data that is collected; 13. an efficacy report is shared with the drug discovery team, in order to improve the drugs; 14. the patient is recoursed with an improved drug; and 2. feedback is sent to the physician for recommendation of a new prescription and dosage. This is one example scenario of the use of BD in the healthcare and pharmaceutical ecosystem. BD has extensive uses, and its potential has still not been fully discovered and practically implemented in real scenarios.

Another scenario based on Figure 1 is shown in Figure 3, which depicts the representation of Scenario 2: Drug discovery for a pandemic situation using BD.

**Figure 3.** Flowchart for drug discovery in pandemic situation using big data.

Scenario 2—Drug discovery for a pandemic situation—uses BD, following the path 3–2–4–6–7–10–6–9–12–14–15–2. First, when a sudden surge of patients with unknown illness is reported at (3.) clinics and (2.) hospitals, (4.) a new set of big data is rapidly created for diagnosis. Prescriptions and pathological reports are generated, (6.) the trends are matched with existing pool to find similar of any historic pandemic; and (7.) the data are quickly modelled to present a pandemic containment prediction predicated on the speed for the contagion vector, as well as its time to mutate, time to outgrow the available medical facilities, and similar data useful for prediction of pandemic containment. Then, the data from 9, 12, and 10 are used recursively used for prediction and drug discovery and provided to (14.) and monitored at (15.) These scenarios represent the application of BD in healthcare and related domains. As data plays the dominant role in the era of data analytics, BD has grea<sup>t</sup> scope in health science and related fields.

Through extensive research, it was confirmed that very few studies have attempted to review existing works to analyse the implementation of BD in the pharmacology, toxicology, and pharmaceutics fields. Thus, the present study is expected to greatly contribute to upcoming research, by providing clear knowledge regarding the implementation of BD in the field of drug development, and will also encourage the development of different big data-based models to serve the medical community. Therefore, the main goal of the present study is to review the literature regarding the employment of BD in pharmaceutics, pharmacology, and toxicology.

*1.1. Objectives of the Study*

> The present study has the following objectives:


## *1.2. Paper Organization*

The introduction section describes the importance of the topic, and the remainder of the article is organised as follows: Section 2 investigates the literature related to the work. Section 3 discusses the benefits of BD. Sections 4 and 5 present the survey analyses. Section 6 presents the limitations of the paper. Finally, Section 7 concludes our survey.

#### *1.3. Definitions and Concepts of BD*

BD can be defined as a huge and complex database that stores data that are heterogeneous and featuring multiple scales of information retrieved from many sources. BD can be related to healthcare in terms of electronic healthcare reports, clinical trial data, and administrative claims. The processing of BD includes feeding the data to the system, data preservation for storage, data analysis, and visualization of outcomes. Grouping the BD and smaller machine resource, offers many advantages, such as high availability, pooling of resources, and scalability [10]. As healthcare information is becoming digitalized, there has been widespread development in healthcare BD, and value-based care has motivated the healthcare sector to use BD analytics to make strategic professional decisions. The healthcare sector is handling many challenges based on the variety, volume, and veracity of data in this sector. Thus, BD plays a huge role by motivating healthcare innovation, influenced by many financial systems, considering factors such as the requirements of patients, motivations of the developers, and technical development. It has been observed that BD can offer optimal treatment decisions for patients and healthcare providers, based on population statistics [11].
