Ontological approach to the knowledge systematization of a toxic process and toxic course representation framework for early drug risk management

Various types of drug toxicity can halt the development of a drug. Because drugs are xenobiotics, they inherently have the potential to cause injury. Clarifying the mechanisms of toxicity to evaluate and manage drug safety during drug development is extremely important. However, toxicity mechanisms, especially hepatotoxic mechanisms, are very complex. The significant exposure of liver cells to drugs can cause dysfunction, cell injury, and organ failure in the liver. To clarify potential risks in drug safety management, it is necessary to systematize knowledge from a consistent viewpoint. In this study, we adopt an ontological approach. Ontology provides a controlled vocabulary for sharing and reusing of various data with a computer-friendly manner. We focus on toxic processes, especially hepatotoxic processes, and construct the toxic process ontology (TXPO). The TXPO systematizes knowledge concerning hepatotoxic courses with consistency and no ambiguity. In our application study, we developed a toxic process interpretable knowledge system (TOXPILOT) to bridge the gaps between basic science and medicine for drug safety management. Using semantic web technology, TOXPILOT supports the interpretation of toxicity mechanisms and provides visualizations of toxic courses with useful information based on ontology. Our system will contribute to various applications for drug safety evaluation and management.


Supplementary Information 1. Toxic course definition and causal relationship representation
In this study, we define toxic courses in TXPO using Protégé 5.2. As described in Methods, the processes specified in each toxic course are shown using 'has part' relation of Object Property. The processes of parent toxic course are inherited by the child toxic courses.
Is-a (subclass of) relationships of processes between toxic courses are represented as follows: For example, by inheriting and specializing 'hypofunction of phospholipid degradation' in the course of phospholipidosis, 'hypofunction of sphingomyelin degradation' in the course of sphingomyelin disorder is defined, which has context 'sphingomyelin disorder' with the constraint owl:allValuesFrom (shown as 'only' in Protégé.) Each process has "has result" relations that describe the possible result process(es) in the toxic course. For example, 'hypofunction of sphingomyelin degradation' has the possible results 'sphingomyelin metabolism imbalance' and 'decreasing ceramide' in sphingomyelin disorder.
The former represents an inherited and specialized process, and the latter represents a new process defined in the child course.

Supplementary Information 2. Toxic imbalance model representation framework
In this imbalance model, we represent the basic units as four processes: 1) a functioning process (supply) as a biological defense for maintaining homeostasis; 2) a functional demand process (demand) as toxic activity; 3) balance/imbalance between toxic activity and defense processes; 4) outcome of the organelles, cells, or tissues of the organ exhibiting toxicity manifestations.
The level of functioning performance can change according to changes in demand; however, if demand exceeds performance, an imbalance and outcomes that manifest toxicity occur. Here, we introduce the following levels: 'very low', 'low', 'moderate', 'high', and 'very high'

1) Normal condition
Since cells normally maintain homeostasis in which the milieu is maintained within a narrow range, we define this level of functioning as 'moderate'.

2) Adaptation
In the body system, if the functional demand increases from normal to a high level, the defense function also performs at a high level and achieves a new homeostasis as adaptation. During the adaptation, it is necessary to maintain homeostasis at a higher state than normal. Examples are hypertrophy and hyperplasia, both of which use processes functioning at a high level to meet increasing demands.

3) Toxicity manifestation from adaptation failures
Even if homeostasis is maintained at a higher level, when the demand increases to 'very high' by some additional factors. As a result, an imbalance can occur between the demand and defense function, which leads to cell injury and death as an outcome.

4) Drug-induced irreversible damage
Under severe stimulus conditions, a high level of drug exposure performs a 'very high' toxic action, which results in an imbalance between the demand and defense function. This results in irreversible cell injury and cell death.

5) Functional excess
Sometimes, the level of biological defense function exceeds demand and becomes 'very high'. In that case, cell damage also occurs. Serious results such as tumor cell proliferation and fibrosis can occur.

6) Lack of adaptation
If the function level cannot be increased from 'moderate' to 'high' for some reason, the ability to adapt will degrade. Accordingly, in response to a stimulus, imbalance and cell damage can occur.
We developed the computer representation framework of the toxic imbalance model and introduced degree levels for representing the functional demand/ supply in TXPO using Protégé 5.2. As illustrated in the original text, the degree of functioning performance can change according to changes in demand; however, if demand exceeds the performance of functioning, an imbalance occurs and results in outcomes that manifest toxicity.
An imbalance/balance process has one toxic composite imbalance state. The toxic composite imbalance state has 1) a biological defense functioning state and 2) a functional demand state, and Both of each defense (supply) and demand have a degree level. The degree level is a subtype of 'qualitative quantity' and has five levels: Very low, small, low, high, and very high in the TXPO is-a hierarchy.
When the normal condition, the balance is keeping represented by 'balance' state supply and functional demand both at Medium levels.
However, in the case of a phospholipid metabolic imbalance (moderate), the imbalance state is represented by a 'toxic imbalance M <H.' 'Toxic imbalance M <H' has: 1) Medium level of the biological defense and 2) High level of functional demand as toxic action.

Supplementary Information 3
Supplementary Figure S1. TOXPIOT system architecture • The RDF data are stored in an RDF triple store using Apache jena Fuseki to construct the SPARQL endpoint. • Regarding the web application system for TOXPILOT, necessary information is dynamically acquired via SPARQL queries • TOXPILOT generates graphs using D3.js of the JavaScript library Supplementary