Article
Biochemical Research Methods
Delora Baptista, Pedro G. Ferreira, Miguel Rocha
Summary: This article critically reviews recent studies that have used deep learning methods to predict drug response in cancer cell lines and introduces the characteristics of DL and the architectures used in these studies. It also provides an overview of publicly available drug screening data resources and discusses the limitations of these methods.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Biochemical Research Methods
Fangfang Xia, Jonathan Allen, Prasanna Balaprakash, Thomas Brettin, Cristina Garcia-Cardona, Austin Clyde, Judith Cohn, James Doroshow, Xiaotian Duan, Veronika Dubinkina, Yvonne Evrard, Ya Ju Fan, Jason Gans, Stewart He, Pinyi Lu, Sergei Maslov, Alexander Partin, Maulik Shukla, Eric Stahlberg, Justin M. Wozniak, Hyunseung Yoo, George Zaki, Yitan Zhu, Rick Stevens
Summary: To enable personalized cancer treatment, machine learning models have been developed to predict drug response based on tumor and drug features. This study used machine learning to analyze five publicly available cell line-based data sets and rigorously evaluated the model generalizability between different studies. The results showed that a multitasking deep neural network achieved the best cross-study generalizability, with models trained on the CTRP data set providing the most accurate predictions on testing data, and the gCSI data set being the most predictable among the cell line data sets.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Bin Liu, Jin Wang, Kaiwei Sun, Grigorios Tsoumakas
Summary: The discovery of drug-target interactions (DTIs) is crucial in pharmaceutical development. Computational methods are promising in predicting novel DTIs. This study presents a Fine-Grained Selective similarity integration approach (FGS) that captures and exploits the importance of similarities at a finer granularity.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Biochemical Research Methods
A. K. M. Azad, Mojdeh Dinarvand, Alireza Nematollahi, Joshua Swift, Louise Lutze-Mann, Fatemeh Vafaee
Summary: Drug similarity studies aim to explore similar therapeutic actions among drugs, utilizing various sources of evidence to derive a multi-modal drug-drug similarity network. The resulting database, DrugSimDB, provides an exhaustive platform for identifying validated repositioning candidates and in-silico drug development applications.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Computer Science, Interdisciplinary Applications
Vijay Kumar, Nitin Dogra
Summary: Drug combination therapies are crucial in cancer treatment, increasing drug efficacy and reducing dosage. The application of deep learning techniques in handling complex synergistic drug combinations is discussed in this paper. Future research should focus on the challenges and directions in the field of drug synergy.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2022)
Article
Cell Biology
Jiacheng Sun, You Lu, Linqian Cui, Qiming Fu, Hongjie Wu, Jianping Chen
Summary: In this paper, a model based on Q-learning algorithm and Neighborhood Regularized Logistic Matrix Factorization (QLNRLMF) is proposed to predict drug-target interactions (DTIs) by fusing heterogeneous information. The model has achieved better effect compared to existing methods in benchmark datasets.
FRONTIERS IN CELL AND DEVELOPMENTAL BIOLOGY
(2022)
Article
Biochemical Research Methods
Ali Akbar Jamali, Anthony Kusalik, Fang-Xiang Wu
Summary: Prediction of drug-target interactions (DTIs) is important in drug development and discovery. Computational DTI prediction is a cost-effective shortcut compared to experimental methods. This study proposes an effective approach, NMTF-DTI, which utilizes multiple kernels and Laplacian regularization to improve prediction performance.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2023)
Article
Automation & Control Systems
Betsabeh Tanoori, Mansoor Zolghadri Jahromi
Summary: The success of drug discovery depends on the identification of drug-target interactions. Computational methods can accelerate this process by predicting binding affinity based on similarity values. This study proposes an effective algorithm for predicting binding affinity using similarity information.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2021)
Article
Biotechnology & Applied Microbiology
Xiao-Ying Liu, Xin-Yue Mei
Summary: With the rapid development of multi-omics technologies and accumulation of large-scale bio-datasets, researchers have conducted comprehensive studies on human diseases and drug sensitivity using multiple biomolecules. However, current drug sensitivity prediction models based on multi-omics data still face challenges such as overfitting and lack of interpretability.
FRONTIERS IN BIOENGINEERING AND BIOTECHNOLOGY
(2023)
Article
Pharmacology & Pharmacy
Xuesong Song, Lin Hou, Yuanyuan Zhao, Qingtian Guan, Zhiwen Li
Summary: In this study, a prognostic signature model for gastric cancer was constructed using 12 metal-dependent programmed cell death-related long noncoding RNAs (lncRNAs), which improved the prediction of overall survival in gastric cancer patients. The analysis also provided guidance for individualized chemotherapy regimen design based on the sensitivity of antitumor drugs.
FRONTIERS IN PHARMACOLOGY
(2022)
Review
Medicine, General & Internal
Alexander Partin, Thomas S. Brettin, Yitan Zhu, Oleksandr Narykov, Austin Clyde, Jamie Overbeek, Rick L. Stevens
Summary: Cancer claims millions of lives yearly worldwide. Exploiting computational predictive models using deep learning methods shows promise in improving drug development and personalized treatment plans for cancer. However, the lack of standardized framework for comparing drug response prediction models hinders the deciphering of predominant and emerging trends.
FRONTIERS IN MEDICINE
(2023)
Article
Biochemical Research Methods
Di He, Lei Xie
Summary: CLEIT is a novel network framework that aims to address challenges in predicting genotype-phenotype associations by integrating multiple incoherent omics data, learning latent representations of high-level domains, and leveraging unlabeled data to improve generalizability of predictive models. It demonstrates effectiveness in predicting anti-cancer drug sensitivity from somatic mutations with the assistance of gene expressions compared to state-of-the-art methods.
Article
Biochemical Research Methods
Stephen Price, Stephane Tombeur, Alexander Hudson, Nanda Kumar Sathiyamoorthy, Paul Smyth, Anjana Singh, Mara Peccianti, Elisa Baroncelli, Ahmed Essaghir, Ilaria Ferlenghi, Sanjay Kumar Phogat, Gurpreet Singh
Summary: Comparing protein structures is crucial for various biological problems, and TMQuery offers a continuously updated database with pre-computed TM-score values for every pair of proteins in the Protein Data Bank, allowing researchers to quickly access these values through a web interface.
Article
Biochemistry & Molecular Biology
Maryam Pouryahya, Jung Hun Oh, James C. Mathews, Zehor Belkhatir, Caroline Moosmueller, Joseph O. Deasy, Allen R. Tannenbaum
Summary: This study proposes a novel network-based methodology to improve the predictive power of anti-cancer drug responses in cell lines. By clustering cell lines and drugs and using gene-expression profiles and drug features for modeling, the accuracy of drug response prediction can be enhanced. The study finds that drugs targeting the PI3K/mTOR signaling pathway perform well in predicting drug responses.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2022)
Article
Genetics & Heredity
Ying Zheng, Zheng Wu
Summary: Drug repositioning is an effective method for predicting drug-target interactions, utilizing heterogeneous networks to construct similarity matrix and employing cascade deep forest method for prediction.
FRONTIERS IN GENETICS
(2021)
Article
Ethics
Kerstin N. Vokinger, Daniel J. Stekhoven, Michael Krauthammer
JOURNAL OF LAW MEDICINE & ETHICS
(2020)
Article
Oncology
Egle Ramelyte, Aizhan Tastanova, Zsolt Balazs, Desislava Ignatova, Patrick Turko, Ulrike Menzel, Emmanuella Guenova, Christian Beisel, Michael Krauthammer, Mitchell Paul Levesque, Reinhard Dummer
Summary: The study found that intralesional T-VEC treatment in primary cutaneous B cell lymphoma patients led to significant tumor response, rapid eradication of malignant cells, activation of the interferon pathway, and early influx of various immune cells, ultimately resulting in enhanced cellular immunity.
Article
Pharmacology & Pharmacy
Isabel Alvarado-Cruz, Mariam Mahmoud, Mohammed Khan, Shilin Zhao, Sebastian Oeck, Rithy Meas, Kaylyn Clairmont, Victoria Quintana, Ying Zhu, Angelo Porciuncula, Hailey Wyatt, Shuangge Ma, Yu Shyr, Yong Kong, Patricia M. LoRusso, Daniel Laverty, Zachary D. Nagel, Kurt A. Schalper, Michael Krauthammer, Joann B. Sweasy
Summary: The study found that long-term treatment with PARP inhibitors induced an inflammatory response in BRCA1 mutant cells, leading to the upregulation of inflammatory genes and activation of the cGAS/STING pathway. In contrast, an increased mutational load was induced in BRCA1-complemented cells treated with a PARP inhibitor.
BIOCHEMICAL PHARMACOLOGY
(2021)
Article
Ethics
Nikola Biller-Andorno, Andrea Ferrario, Susanne Joebges, Tanja Krones, Federico Massini, Phyllis Barth, Georgios Arampatzis, Michael Krauthammer
Summary: This paper focuses on the ethical questions around the design, development and deployment of AI systems in decision-making processes related to cardiopulmonary resuscitation and code status determination in healthcare. It discusses the challenges in current practices and the openness among healthcare professionals to consider AI-based decision support. The data suggest potential for AI to improve decision-making in resuscitation, with a set of ethically relevant preconditions needing consideration for further development and implementation efforts.
JOURNAL OF MEDICAL ETHICS
(2022)
Article
Biochemical Research Methods
Kyriakos Schwarz, Ahmed Allam, Nicolas Andres Perez Gonzalez, Michael Krauthammer
Summary: The study introduces a Siamese self-attention multi-modal neural network for predicting drug-drug interactions (DDIs), demonstrating accurate predictions and improved performance, as well as interpretability through an Attention mechanism.
BMC BIOINFORMATICS
(2021)
Article
Microbiology
Ana Montalban-Arques, Egle Katkeviciute, Philipp Busenhart, Anna Bircher, Jakob Wirbel, Georg Zeller, Yasser Morsy, Lubor Borsig, Jesus F. Glaus Garzon, Anne Mueller, Isabelle C. Arnold, Mariela Artola-Boran, Michael Krauthammer, Anna Sintsova, Nicola Zamboni, Gabriel E. Leventhal, Laura Berchtold, Tomas de Wouters, Gerhard Rogler, Katharina Baebler, Marlene Schwarzfischer, Larissa Hering, Ivan Olivares-Rivas, Kirstin Atrott, Claudia Gottier, Silvia Lang, Onur Boyman, Ralph Fritsch, Markus G. Manz, Marianne R. Spalinger, Michael Scharl
Summary: Despite the overall success of T cell checkpoint inhibitors in cancer treatment, research has found that certain commensal species of gut microbiota associated with lower tumor burden and are significantly reduced in CRC patients. Oral application of specific Clostridiales strains can prevent and even treat CRC successfully in mouse models, outperforming anti-PD-1 therapy in some cases. This indicates the potential of gut bacteria as a novel stand-alone therapy against solid tumors.
CELL HOST & MICROBE
(2021)
Article
Multidisciplinary Sciences
Kim F. Marquart, Ahmed Allam, Sharan Janjuha, Anna Sintsova, Lukas Villiger, Nina Frey, Michael Krauthammer, Gerald Schwank
Summary: Base editors, consisting of a CRISPR-Cas module and a DNA deaminase, allow for precise genetic alterations by converting C∙G to T∙A base pairs and vice versa. However, their editing efficiencies vary across different genomic loci. Researchers conducted an analysis on over 28,000 integrated genetic sequences to develop BE-DICT, a machine learning model capable of predicting base editing outcomes accurately.
NATURE COMMUNICATIONS
(2021)
Article
Medicine, General & Internal
Martina A. Maibach, Ahmed Allam, Matthias P. Hilty, Nicolas A. Perez Gonzalez, Philipp K. Buehler, Pedro D. Wendel Garcia, Silvio D. Brugger, Christoph C. Ganter, Michael Krauthammer, Reto A. Schuepbach, Jan Bartussek
Summary: The study found that aligning patient data using CRP peak values can improve mortality risk stratification in COVID-19 patient cohorts. The data highlights the importance of proper synchronization of longitudinal patient data for accurate interpatient comparisons and relevant subgroup definitions. The use of objective temporal disease markers can facilitate translational research efforts and multicenter trials.
FRONTIERS IN MEDICINE
(2021)
Article
Health Care Sciences & Services
Ahmed Allam, Stefan Feuerriegel, Michael Rebhan, Michael Krauthammer
Summary: Patient trajectories in digital medicine, which record health events over time, can predict the future course of diseases. However, current AI solutions need to adapt to analyze the rich longitudinal data in order to build robust personalized models for risk scoring and disease pathway discovery.
JOURNAL OF MEDICAL INTERNET RESEARCH
(2021)
Article
Medicine, General & Internal
Uwe Bieri, Michael Scharl, Silvan Sigg, Barbara Maria Szczerba, Yasser Morsy, Jan Hendrik Ruschoff, Peter Hans Schraml, Michael Krauthammer, Lukas John Hefermehl, Daniel Eberli, Cedric Poyet
Summary: The human microbiota has been linked to inflammatory and neoplastic diseases. While the gut microbiome has been extensively studied, the urinary microbiome is still relatively new. Investigating the relationship between bladder cancer and the bladder and intestinal microbiome may provide insights into their pathophysiological relationship. It may also lead to the discovery of non-invasive biomarkers for tumor behavior.
Article
Biology
Gaetana Restivo, Aizhan Tastanova, Zsolt Balazs, Federica Panebianco, Maren Diepenbruck, Caner Ercan, Bodgan-T Preca, Juerg Hafner, Walter P. Weber, Christian Kurzeder, Marcus Vetter, Simone Munst Soysal, Christian Beisel, Mohamed Bentires-Alj, Salvatore Piscuoglio, Michael Krauthammer, Mitchell P. Levesque
Summary: Fresh and slow-frozen tissues from various malignancies are comparable and suitable for different methods, including culture establishment and RNA sequencing, with similar results to freshly analyzed material.
COMMUNICATIONS BIOLOGY
(2022)
Article
Biotechnology & Applied Microbiology
Nicolas Mathis, Ahmed Allam, Lucas Kissling, Kim Fabiano Marquart, Lukas Schmidheini, Cristina Solari, Zsolt Balazs, Michael Krauthammer, Gerald Schwank
Summary: This study conducted a high-throughput screen to identify sequence context features that influence prime editing outcomes. They trained a neural network called PRIDICT, which reliably predicts editing rates for various genetic changes. The validation results showed that PRIDICT can significantly increase prime editing efficiencies in different cell types.
NATURE BIOTECHNOLOGY
(2023)
Article
Rheumatology
Alexandru Garaiman, Farhad Nooralahzadeh, Carina Mihai, Nicolas Perez Gonzalez, Nikitas Gkikopoulos, Mike Oliver Becker, Oliver Distler, Michael Krauthammer, Britta Maurer
Summary: The study aims to evaluate the performance of a vision transformer-based deep-learning model in identifying microangiopathy in NFC images of SSc patients and compare it with the performance of rheumatologists. The results show that the ViT performs well in identifying different microangiopathic changes, but there is still room for improvement in terms of reliability compared to rheumatologists.
Article
Ethics
Jeffrey David Iqbal, Michael Krauthammer, Nikola Biller-Andorno
Summary: This paper introduces the concept and use cases of digital twins in medicine, discusses ethical challenges, and explores their impact on future medical practice.
JOURNAL OF LAW MEDICINE & ETHICS
(2022)
Article
Biochemical Research Methods
Aizhan Tastanova, Egle Ramelyte, Zsolt Balazs, Ulrike Menzel, Christian Beisel, Michael Krauthammer, Reinhard Dummer, Mitchell Paul Levesque
Summary: High cell viability and recovered cell concentration are crucial for quality control in single-cell processing and data analysis. This protocol details procedures for sampling, preprocessing, and analyzing fine-needle aspiration (FNA) samples of the skin, which is minimally invasive and allows for longitudinal sampling to capture cellular heterogeneity in clinical samples.