Article
Biochemical Research Methods
Deniz Secilmis, Thomas Hillerton, Sven Nelander, Erik L. L. Sonnhammer
Summary: The study introduces an algorithm called IDEMAX to infer effective perturbation design from gene expression data, improving the accuracy of GRN inference in real data where noise often hides much of the signal.
Review
Biochemical Research Methods
Mengyuan Zhao, Wenying He, Jijun Tang, Quan Zou, Fei Guo
Summary: The study focuses on the importance of GRN reconstruction technologies in biology and medical science, discussing different method classifications and their performance in networks of varying scales. The aim is to discover potential drug targets and identify cancer biomarkers.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Biochemical Research Methods
Mehrzad Saremi, Maryam Amirmazlaghani
Summary: In this study, an algorithm called GENEREF was developed to accumulate information from multiple datasets in an iterative manner, improving the accuracy of prediction results. The algorithm was extensively tested on multiple datasets and outperformed other algorithms on selected networks, showing competitiveness with existing multi-dataset algorithms.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2022)
Article
Biochemical Research Methods
Cassandra Burdziak, Chujun Julia Zhao, Doron Haviv, Direna Alonso-Curbelo, Scott W. Lowe, Dana Pe'er
Summary: scKINETICS is a dynamical model that fits gene expression change with the learning of per-cell transcriptional velocities and a governing gene regulatory network. It successfully recapitulates the process of acinar-to-ductal transdifferentiation and proposes novel regulators of this process in an acute pancreatitis dataset. In benchmarking experiments, scKINETICS extends and improves existing velocity approaches to generate interpretable, mechanistic models of gene regulatory dynamics.
Article
Biochemical Research Methods
Wenying He, Jijun Tang, Quan Zou, Fei Guo
Summary: Gene regulatory networks (GRNs) play a crucial role in biological processes, and our proposed method MMFGRN shows promising results in reconstructing these networks by leveraging different data types to explore potential regulatory relationships.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Genetics & Heredity
Juan M. Escorcia-Rodriguez, Estefani Gaytan-Nunez, Ericka M. Hernandez-Benitez, Andrea Zorro-Aranda, Marco A. Tello-Palencia, Julio A. Freyre-Gonzalez
Summary: Gene regulatory networks are models representing cellular transcription events. Time and resource constraints make them far from complete. Previous assessments have shown the limitations of methods for inferring these networks. This study highlights the importance of data quality, assessment approach, and network structure for accurate network inference.
FRONTIERS IN GENETICS
(2023)
Article
Biochemical Research Methods
Julia Akesson, Zelmina Lubovac-Pilav, Rasmus Magnusson, Mika Gustafsson
Summary: ComHub is a tool for predicting hubs in GRNs by averaging predictions from a compendium of network inference methods. Benchmarking against DREAM5 challenge data and independent gene expression datasets showed robust performance. ComHub consistently scored among the top performing methods on data from different sources, and can work with both predefined networks and perform stand-alone network inference, making it generally applicable.
BMC BIOINFORMATICS
(2021)
Article
Physics, Multidisciplinary
Taylor Firman, Jonathan Huihui, Austin R. Clark, Kingshuk Ghosh
Summary: Researchers have found that inference based on the Maximum Caliber principle is more efficient in extracting hidden information from single-cell stochastic gene expression time trajectories, compared to traditional modeling methods, demonstrating greater accuracy and computational efficiency.
Article
Biochemistry & Molecular Biology
Yongqing Zhang, Maocheng Wang, Zixuan Wang, Yuhang Liu, Shuwen Xiong, Quan Zou
Summary: In this study, a meta-learning framework for gene regulatory network (GRN) inference was proposed to identify cell states and important regulators. The framework solved the parameter optimization problem caused by high-dimensional sparse data features using meta-learning, and addressed the lack of label data through few-shot learning. A structural equation model was embedded in the model to identify important regulators. By studying the GRN inference task and cell type identification, it was confirmed that the selected regulators were closely related to gene expression specificity. Extensive experimental results demonstrated the effectiveness of the model in single-cell GRN inference, and the visualization results verified the importance of the selected regulators for cell type recognition.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Review
Mathematics, Applied
Manuela A. D. Aguiar, Ana P. S. Dias, Haibo Ruan
Summary: This paper discusses the synchronization mechanisms in GRNs as well as the impact of gene duplication and redundancy on them, presenting results on robust synchrony patterns for SUM and MULT dynamical models. It also explores the concepts of quotient network and network lifting, and their relationship to gene duplication and the phenomenon of functional divergence.
PHYSICA D-NONLINEAR PHENOMENA
(2022)
Article
Mathematics
Wenlong He, Peng Xia, Xinan Zhang, Tianhai Tian
Summary: The rapid progress in biological experimental technologies has led to a wealth of experimental data for studying complex regulatory mechanisms. Mathematical models have been proposed to simulate the dynamics of molecular processes, but estimating unknown parameters in these models for different cells remains challenging due to the computational demands. In this study, a population statistical inference algorithm is proposed to improve computational efficiency and infer parameters of regulatory networks in a large number of cells.
Article
Biology
Adrian Segura-Ortiz, Jose Garcia-Nieto, Jose F. Aldana-Montes, Ismael Navas-Delgado
Summary: Gene regulatory networks play a crucial role in understanding disease triggers and developing new therapeutic targets. This study proposes GENECI, an evolutionary machine learning approach, to construct ensembles of inference results and optimize the consensus network based on confidence levels and topological characteristics. The proposed method is proven to be robust and accurate, with the ability to generalize to multiple datasets.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Biochemical Research Methods
Juan D. Henao, Michael Lauber, Manuel Azevedo, Anastasiia Grekova, Fabian Theis, Markus List, Christoph Ogris, Benjamin Schubert
Summary: This study integrated regression-based methods that can handle missingness into KiMONo, and benchmarked their performance on commonly encountered missing data scenarios in single- and multi-omics studies. The results showed that two-step approaches that explicitly handle missingness performed best for imbalanced omics-layers dimensions, while methods implicitly handling missingness performed best for balanced omics-layers dimensions. The study demonstrated the feasibility of robust multi-omics network inference in the presence of missing data with KiMONo.
BRIEFINGS IN BIOINFORMATICS
(2023)
Article
Multidisciplinary Sciences
David A. Rand, Archishman Raju, Meritxell Saez, Francis Corson, Eric D. Siggia
Summary: Embryonic development leads to reproducible and ordered complexity from egg to adult, with different cell types differentiating due to gene networks. Geometric methods formalize such landscapes, while Smale's results suggest gene network systems can be represented as potential gradients. By tuning parameters, all cell states vs. model parameters can be enumerated.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Oncology
Aleksandar Obradovic, Casey Ager, Mikko Turunen, Thomas Nirschl, Mohsen Khosravi-Maharlooei, Alina Iuga, Christopher M. Jackson, Srinivasan Yegnasubramanian, Lorenzo Tomassoni, Ester Calvo Fernandez, Patrick McCann, Meri Rogava, Angelo M. DeMarzo, Christina M. Kochel, Mohamad Allaf, Trinity Bivalacqua, Michael Lim, Ronald Realubit, Charles Karan, Charles G. Drake, Andrea Califano
Summary: By analyzing the transcriptional state of tumor-infiltrating regulatory T cells (TI-Tregs), 17 candidate master regulators (MRs) were identified as important determinants in TI-Treg recruitment and retention. CRISPR-Cas9 screening confirmed the essentiality of 8 MRs in TI-Treg function and targeting one of the most significant MRs (Trps1) successfully reduced tumor growth. Additionally, low-dose gemcitabine was found to reverse the activity of TI-Treg MRs, leading to tumor inhibition.
Article
Computer Science, Information Systems
Rashmi Gupta, Martin Crane, Cathal Gurrin
Summary: The study found that lifeloggers tend to prefer conservative privacy settings and have varying willingness to share different categories of images. Participants preferred to keep personally identifiable information and professional information private, while being open to sharing family moments or daily routine content with family/friends, and potentially making other visual lifelog data public.
ONLINE INFORMATION REVIEW
(2021)
Article
Biology
Andreas Hillmann, Martin Crane, Heather J. Ruskin
JOURNAL OF THEORETICAL BIOLOGY
(2020)
Article
Multidisciplinary Sciences
Alina Sirbu, Greta Barbieri, Francesco Faita, Paolo Ferragina, Luna Gargani, Lorenzo Ghiadoni, Corrado Priami
Summary: Clinical decision support systems based on predictive models can improve the management of COVID-19 patients by selecting six clinical variables with the highest predictive power using genetic algorithms.
SCIENTIFIC REPORTS
(2021)
Correction
Computer Science, Artificial Intelligence
Alina Sirbu, Gennady Andrienko, Natalia Andrienko, Chiara Boldrini, Marco Conti, Fosca Giannotti, Riccardo Guidotti, Simone Bertoli, Jisu Kim, Cristina Ioana Muntean, Luca Pappalardo, Andrea Passarella, Dino Pedreschi, Laura Pollacci, Francesca Pratesi, Rajesh Sharma
INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS
(2021)
Article
Computer Science, Theory & Methods
Tai Tan Mai, Marija Bezbradica, Martin Crane
Summary: With the shift of higher education programmes to online channels due to the COVID19 pandemic, issues in monitoring students' learning progress have arisen. However, a novel approach has been proposed to analyze students' learning behavior and its relationship with learning assessment results, improving the analysis and prediction based on learning behavioral datasets.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2022)
Article
Mathematics, Applied
Jaroslaw Kwapien, Marcin Watorek, Marija Bezbradica, Martin Crane, TaiTan Mai, Stanislaw Drozdz
Summary: This study analyzes tick-by-tick data of major cryptocurrencies traded on different cryptocurrency trading platforms, examining quantities such as inter-transaction times, number of transactions, traded volume, and volatility. The study shows that inter-transaction times exhibit long-range power-law autocorrelations, with periods of increased market activity displaying richer multifractality compared to quiet periods. Additionally, neither stretched exponential distribution nor power-law-tail distribution universally models the cumulative distribution functions of the examined quantities. It is also observed that parallel data sets from different trading platforms can have disparate statistical properties.
Article
Physics, Multidisciplinary
An Pham Ngoc Nguyen, Tai Tan Mai, Marija Bezbradica, Martin Crane
Summary: This study analyzes the correlation between different assets in the cryptocurrency market and examines investment decisions during bearish and bullish periods. By utilizing fine-grained time series data and community detection algorithms, the study reveals clearer correlations among cryptocurrencies and proposes a noise and trend removal scheme for the original correlations.
Review
Mathematics, Interdisciplinary Applications
Jisu Kim, Alina Sirbu, Fosca Giannotti, Giulio Rossetti, Hillel Rapoport
Summary: The cultural integration of immigrants has a significant impact on their overall socio-economic integration and the attitudes of natives towards globalization and immigration. Cultural integration can be described as the preservation of ties to the origin country and culture, as well as the creation of new links and adoption of cultural traits from the new residence country. This paper introduces a method based on Twitter data to quantify these aspects and analyzes their possible determinants.
Article
Mathematics
Conall Butler, Martin Crane
Summary: This paper investigates the transaction dynamics of the Ethereum network after the London Hard Fork and explores the relationship between Ethereum price and gas price. It compares different machine learning methods for gas price forecasting and finds that hybrid models perform better than attention and CNN-LSTM models due to hardware constraints. By providing forecasts for different lookaheads, users can make more informed decisions on gas price selection and transaction submission timing.
Article
Physics, Multidisciplinary
Tai Tan Mai, Martin Crane, Marija Bezbradica
Summary: The high dropout rates in programming courses necessitate monitoring and understanding of student engagement for early interventions. This study introduces entropy-based metrics as a novel way to represent students' learning behaviors and analyzes the behaviors of higher- and lower-performing student communities using a proven community detection method. The impact of the COVID-19 pandemic on these behaviors is also examined. The findings demonstrate the value of using entropy as a simple yet insightful metric for educators to monitor study progress, enhance understanding of student engagement, and enable timely interventions.
Article
Computer Science, Information Systems
Jisu Kim, Francesca Pratesi, Giulio Rossetti, Alina Sirbu, Fosca Giannotti
Summary: This study fills the gap in previous research on social networks by analyzing the characteristics and behaviors of migrants and natives on Twitter. It found that migrants have more followers and tweet more compared to natives, and their connections are more based on nationality. Additionally, both groups exhibit strong homophilic behaviors.
SOCIAL NETWORK ANALYSIS AND MINING
(2022)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Jisu Kim, Alina Sirbu, Giulio Rossetti, Fosca Giannotti
Summary: This paper fills a gap in the research on social networks by studying the characteristics and behaviors of migrants and natives on Twitter. The study finds that migrants have more followers and tweet more frequently. Additionally, users tend to connect based on nationality rather than country of residence, and this is more pronounced for migrants.
COMPLEX NETWORKS & THEIR APPLICATIONS X, VOL 1
(2022)
Article
Multidisciplinary Sciences
Eoin Cartwright, Martin Crane, Heather J. Ruskin
Summary: This article discusses a novel approach to motif identification that allows for flexibility in side-length, improving the recognition of localized similar behavior across varying timescales. It is of significance to applications in the financial and energy sectors.
Proceedings Paper
Education & Educational Research
Tai Tan Mai, Martin Crane, Marija Bezbradica
Summary: This article analyzes the learning behaviors of students in an introductory programming course at a University in Dublin, finding that high-performing students are active in practice while low-performing students tend to focus on reading lecture notes and show signs of discouragement and lack of motivation in practical activities towards the end of the semester.
7TH INTERNATIONAL CONFERENCE ON HIGHER EDUCATION ADVANCES (HEAD'21)
(2021)
Article
Computer Science, Artificial Intelligence
Alina Sirbu, Gennady Andrienko, Natalia Andrienko, Chiara Boldrini, Marco Conti, Fosca Giannotti, Riccardo Guidotti, Simone Bertoli, Jisu Kim, Cristina Ioana Muntean, Luca Pappalardo, Andrea Passarella, Dino Pedreschi, Laura Pollacci, Francesca Pratesi, Rajesh Sharma
Summary: This paper explores how big data can assist in understanding the migration phenomenon, focusing on the journey, stay, and effects of migration on source countries. Traditional and novel data sources and models are compared across different phases of migration, aiming to provide a basis for a novel multi-level integration index.
INTERNATIONAL JOURNAL OF DATA SCIENCE AND ANALYTICS
(2021)