약물의 염전성 부정맥 유발 예측 지표로서 심장의 전기생리학적 특징 값들의 검증

유예담,정다운,Aroli Marcellinus,임기무 대한의용생체공학회 2022 의공학회지 Vol.43 No.1

The Comprehensive in vitro Proarrhythmic Assay(CiPA) project was launched for solving the hERG assay problem of being classified as high-risk groups even though they are low-risk drugs due to their high sensitivity. CiPA presented a protocol to predict drug toxicity using physiological data calculated based on the in-silico model. in this study, features calculated through the in-silico model are analyzed for correlation of changing action potential in the near future, and features are verified through predictive performance according to drug datasets. Using the O'Hara Rudy model modified by Dutta et al., Pearson correlation analysis was performed between 13 features(dVm/dtmax, APpeak, APresting, APD90, APD50, APDtri, Capeak, Caresting, CaD90, CaD50, CaDtri, qNet, qInward) calculated at 100 pacing, and between dVm/dtmax_repol calculated at 1,000 pacing, and linear regression analysis was performed on each of the 12 training drugs, 16 verification drugs, and 28 drugs. Indicators showing high coefficient of determination(R2) in the training drug dataset were qNet 0.93, AP resting 0.83, APDtri 0.78, Ca resting 0.76, dVm/dtmax 0.63, and APD90 0.61. The indicators showing high determinants in the validated drug dataset were APDtri 0.94, APD90 0.92, APD50 0.85, CaD50 0.84, qNet 0.76, and CaD90 0.64. Indicators with high coefficients of determination for all 28 drugs are qNet 0.78, APD90 0.74, and qInward 0.59. The indicators vary in predictive performance depending on the drug data- set, and qNet showed the same high performance of 0.7 or more on the training drug dataset, the verified drug data- set, and the entire drug dataset.

기계학습을 활용한 약물의 심장독성 위험군 분류

유예담,임기무 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.8

심근 세포의 전기생리학적 특징을 이용한 인공 신경망 기반 약물의 심장독성 평가

유예담,정다운,임기무,Yoo, Yedam,Jeong, Da Un,Marcellinus, Aroli,Lim, Ki Moo 대한의용생체공학회 2021 의공학회지 Vol.42 No.6

Cardiotoxicity assessment of all drugs has been performed according to the ICH guidelines since 2005. Non-clinical evaluation S7B has focused on the hERG assay, which has a low specificity problem. The comprehensive in vitro proarrhythmia assay (CiPA) project was initiated to correct this problem, which presented a model for classifying the Torsade de pointes (TdP)-induced risk of drugs as biomarkers calculated through an in silico ventricular model. In this study, we propose a TdP-induced risk group classifier of artificial neural network (ANN)-based. The model was trained with 12 drugs and tested with 16 drugs. The ANN model was performed according to nine features, seven features, five features as an individual ANN model input, and the model with the highest performance was selected and compared with the classification performance of the qNet input logistic regression model. When the five features model was used, the results were AUC 0.93 in the high-risk group, AUC 0.73 in the intermediate-risk group, and 0.92 in the low-risk group. The model's performance using qNet was lower than the ANN model in the high-risk group by 17.6% and in the low-risk group by 29.5%. This study was able to express performance in the three risk groups, and it is a model that solved the problem of low specificity, which is the problem of hERG assay.

이온채널 전류특성이 TdP 발생에 미치는 영향에 대한 민감도 분석

유예담(Yedam Yoo),임기무(Ki Moo Lim) 대한기계학회 2021 대한기계학회 춘추학술대회 Vol.2021 No.4

CiPA (The comprehensive in vitro pro-arrhythmia assay) 패러다임은 in silico 모델을 이용한 약물의 심장독성 평가방식을 제안하였다. 패러다임에서는 Dutta 모델을 통해 계산된 qNet을 TdP (torsades de pontes)의 발생 가능성 지표로써 사용하였다. 본 연구에서는 EAD (early after depolarizations) 발생에 영향을 주는 이온채널 전도도의 민감도와, qNet에 영향을 주는 이온채널 전도도의 민감도를 분석하였다. 그리고 추가적으로 qNet 지표와 TdP발생과의 상관성을 수치적으로 규명하였다. 수정된 ORD 모델의 이온채널 6 개(IKr, IKs, INaL, ICaL, IK1, Ito)의 채널 막힘 정도를 0%, 25%, 50%, 75%로 선형적으로 변화를 주어 4,096 개의 시나리오를 계산하였다. 채널 민감도 분석은 이온전도도 시나리오와 계산된 값을 입력으로 pearson 상관성 분석을 통해 영향력 순위를 확인하였다. TdP 발생 양상은 GKr 가 -0.94, GNaL 은 0.22의 상관계수를 가졌다. qNet 은 GNaL 이 -0.67, GKr 이 0.65 로 상관성을 가졌다. GKr 은 TdP 발생에 가장 큰 영향을 미쳤으며 qNet 에서는 GKr 과 GNaL 이 주로 영향을 미쳤다. 또한 qNet 과 TdP 발생 사이의 상관계수는 -0.80 로 높은 음의 상관관계를 가졌다. 따라서 qNet 은 통계적인 관점에서 TdP 예측의 유의미한 지표임을 확인하였다.

심근세포의 활동전위로부터 약물의 심장 독성 위험군을 분류하기 위한 함성곱 신경망의 활용

유예담(Yedam Yoo),임기무(Ki Moo Lim) 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2

약물로 인해 발생하는 심장의 부작용은 1960 년대부터 발생되어왔다. 약물로 인해 발생되는 심장 독성 부작용을 미연에 방지하고자 2005 년 시장에 유통되기전 시험하는 심장독성 평가 가이드라인을 만들어 규제하고 있다. 하지만 기존의 약물 독성평가를 개정하고자 새로운 복합적 부정맥 유도 평가(CiPA)를 2013 년에 시작하였다. 이러한 새로운 패러다임은 약물의 독성평가를 시험관내 시험에서 컴퓨터 모델링을 통한 시험이다. 본 연구에서는 컴퓨터 심근 세포모델을 통해 계산된 활동전위 형상을 입력으로 하여 추출된 특성으로 일차원 합성곱 인공신경망 모델을 사용한다. 약물의 위험군은 부정맥 유발정도에 따라 고 위험군, 중위험군, 저위험군으로 분류한다. 패러다임에 따른 불확실성 정량화를 따르고 계산된 약물의 위험군 분류 성능은 AUC 를 기준으로 고위험군은 0.97, 중위험군은 0.8, 저위험군은 0.96 의 성능을 보였다.

인공 신경망을 이용한 in silico 바이오 마커 기반 약물의 부정맥 유발 위험도 분류

유예담(Ye Dam Yoo),임기무(Ki Moo Lim) 대한기계학회 2020 대한기계학회 춘추학술대회 Vol.2020 No.12

다형 심실빈맥의 예측을 위한 dVm/dt_{Max_repol}의 이온채널 전도도에 대한 민감도 분석

정다운,유예담,임기무,Jeong, Da Un,Yoo, Yedam,Marcellinus, Aroli,Lim, Ki Moo 대한의용생체공학회 2022 의공학회지 Vol.43 No.5

Early afterdepolarization (EAD), a significant cause of fatal ventricular arrhythmias including Torsade de Pointes (TdP) in long QT syndromes, is a depolarizing afterpotential at the plateau or repolarization phase in action potential (AP) profile early before completing one pace. AP duration prolongation is related to EAD but is not necessarily accounted for EAD. Several computational studies suggested EAD can form from an abnormality in the late plateau and/or repolarization phase of AP shape. In this sense, we hypothesized the slope during repolarization has the characteristics to predict TdP risk, mainly focusing on the maximum slope during repolarization (dVm/dt_{max_repol}). This study aimed to predict the sensitivity of dVm/dt_{max_repol} to ion channel conductances as a TdP risk metric through a population simulation considering multiple effects of simultaneous reduction in six ion channel conductances of g_NaL, g_Kr, g_Ks, g_to, g_K1, and g_CaL. Additionally, we verified the availability of dVm/dt_{max_repol} for TdP risk prediction through the correlation analysis with qNet, the representative TdP metric. We performed the population simulations based on the methodology of Gemmel et al. using the human ventricular myocyte model of Dutta et al. Among the sixion channel conductances, dVm/dt_{max_repol} and qNet responded most sensitively to the change in g_Kr, followed by g_NaL. Furthermore, dVm/dt_{max_repol} showed a statistically significant high negative correlation with qNet. The dVm/dt_{max_repol} values were significantly different according to three TdP risk levels of high, intermediate, and low by qNet (p<0.001). In conclusion, we suggested dVm/dt_{max_repol} as a new biomarker metric for TdP risk assessment.

Assessment of Drug-Induced Cardiac Arrhythmicity using Deep Learning Approach

Nurul Qashri Mahardika T,Yedam Yoo(유예담),Aroli Marcelinus,Ki Moo Lim(임기무) 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2

One of the considering things about drug development is some drugs can cause Torsade de Pointes (TdP) which is harmful to society. Thus, the Comprehensive in vitro proarrhythmia assay (CiPA) project was successful in classifying to assess drug-induced arrhythmias through a logistic regression model using qNet as a TdP risk assessment biomarker. However, obtaining that result through in silico simulation is inefficient. We propose using the CNN model to classify three levels of proarrhythmic risk: high, intermediate, and low. To obtain net charges and calcium transients, we use the ToRd myocyte cell model. Drug effects were implemented using IC50 and Hill coefficient. There are 12 drugs as CNN model training data and 16 drugs as test data. qNet variability, qInward variability, CaD90 variability, CaD50 variability, and Ca resting variability are used as input in the CNN model independently. As the result, qInward variability showed better performance for the three classifications, 83% high-risk drug, 80% intermediate, and 80% low-risk drug.

Comprehensive in vitro Proarrhythmia Assay (CiPA) 기반 Torsade Metric Score (TMS)를 이용한 TdP 발생 약물의 위험군 분류 성능 평가

윤승현 ( Seung-hyun Yoon ),남유리 ( Yu Ri Nam ),강현수 ( Hyun-su Kang ),정다운 ( Da Un Jeong ),유예담 ( Ye Dam Yoo ),임기무 ( Ki-moo Lim ),박성준 ( Seong-jun Park ),김배환 ( Bae-hwan Kim ),김기석 ( Ki-suk Kim ) 한국동물실험대체법학회 2021 동물실험대체법학회지 Vol.15 No.1

For the assessment of cardiotoxicity and Torsade de Pointes (TdP), a fatal arrhythmic symptom, ICH S7B and E14 guidelines were presented. However, focusing on hERG block, which are essential determinants of arrhythmic risk, may unexpectedly limit drug development by increasing the risk of drugs that are actually non-toxic. To compensate for these problems, the Comprehensive in vitro Proarrhythmic Assay (CiPA) project was proposed. In this study, based on the CiPA project and previous studies, the nine drugs were tested using in vitro multiple ion channel screening on both temperature conditions (room temperature and 37℃). Using the in vitro results, in silico computer simulation was performed based on the O'Hara-Rudy human ventricular myocyte model, and same as the CiPA project obtained a new biomarker, qNet. The in silico computer simulation was performed using 2000 samples of IC₅₀ values extracted by R code. The nine test drugs were associated with cardiotoxicity and TdP, and were selected by the CiPA project and previous studies. Furthermore, as in previous studies, Torsade Metic Score (TMS), the mean qNet value averaged across 1-4 × Cmax, and the threshold was calculated. As a result, the nine tested drugs using the TMS were well plotted by the risk categories and the threshold was able to well classify the risk categories by grade on both temperature conditions. In particular, the threshold 2 value confirmed to significant difference depending on the temperature conditions. The range of threshold narrowed at 37℃, which can be considered as having the ability to distinguish more finely. It shows the correlation with the CiPA project’s validity that a study should be tested at a physiological temperature of 3 7℃. In this study, using the method proposed by the CiPA project, it was possible to predict the risk groups of drugs more accurately, which could be presented as a new paradigm in the cardiotoxicity assessment.