ML-Bioinfo-CEITEC

News from the #MachineLearning in #Bioinformatics lab at @CEITEC_Brno headed by Panagiotis Alexiou - @PanosAlexiou01. #CEITEC #MasarykUniversity

[ML-Bioinfo-CEITEC @MCeitec]

Our new study in using Convolutional Neural Networks to classify G-quadruplexes with increased sample category imbalance.

https://t.co/rrQDQG1imN

[ML-Bioinfo-CEITEC @MCeitec]

Our review paper has been published in @Biology_MDPI .

[Biology MDPI @Biology_MDPI]

Bioinformatics and Machine Learning Approaches to Understand the Regulation of Mobile Genetic Elements https://t.co/160v3l6YJA # via @Biology_MDPI @MasarykUni

#transposable_elements_regulation #mobile_genetic_elements #machine_learning #bioinformatics_methods #DNA_methylation

[CEITEC @CEITEC_Brno]

A unique #bioinformatics tool MuStARD which can scan large sets of genomic regions & precisely identify areas producing small RNAs playing various roles in the regulation of development & disease. It is a novel tool designed by @MCeitec. @SciReports
https://t.co/u2Nql1XCkM

[ML-Bioinfo-CEITEC @MCeitec]

@TDataScience presents Model Compression: needs and importance written by Sabina Pokhrel:
https://t.co/J5XC1HSSfJ

[ML-Bioinfo-CEITEC @MCeitec]

#ChineseAcademyofSciences #KingAbdulazizUniversity used GraRep embedding model to predict miRNA-disease associations; a heterogeneous information network was constructed by combining the known associations among lncRNA, drug, protein, disease, and miRNA. https://t.co/PsLUe6cFA0

[ML-Bioinfo-CEITEC @MCeitec]

@CSU_China @usask @ODU used singular value decomposition and deep learning to extract linear and non-linear features of lncRNAs and diseases, respectively and trained SDLDA by combing the linear and non-linear features. https://t.co/wwmVHsBOOW

[ML-Bioinfo-CEITEC @MCeitec]

@ualbany @urichmond present framework determining validity of sample robot configuration through combination of CAE (occupancy grids representation of the robot's workspace) and Multilayer Perceptron (collision state of robot and robot's configuration). https://t.co/44uJJmMyyu

[ML-Bioinfo-CEITEC @MCeitec]

@VIT_univ discerns drug-responsive coherent genetic markers of OCD, and present a filter-ensemble fused feature selection model. Furthermore, to improve the predictability of the learning model, an unsupervised #DL-based feature extraction method is used. https://t.co/e94a5tyLWH

[ML-Bioinfo-CEITEC @MCeitec]

Top 20 Machine Learning & Data Science Websites To Follow in 2020 written by @OKharkovyna in @TDataScience

https://t.co/i4OWC4BCQP

[ML-Bioinfo-CEITEC @MCeitec]

@CUMT1909 presented a neoteric Bayesian model (KBMFMDA) that combines kernel-based nonlinear dimensionality reduction, matrix factorization, and binary classification to estimate the association network of miRNAs and human complex diseases in a subspace. https://t.co/xOUQNmEaAR

[ML-Bioinfo-CEITEC @MCeitec]

@SUSTechSZ @Akademie_ved_CR proposed new method DeepTopPush for minimizing the top loss function. DeepTopPush had good performance on visual recognition datasets and on real-world application of selecting small number of molecules for further drug testing. https://t.co/XYyPQ4NqIS

[ML-Bioinfo-CEITEC @MCeitec]

@uarizona @UAZHealth assess the classification ability, reproducibility of statistical learning tools for gene biomarker detection with six state-of-the-art #ML models and demonstrate the challenges of reproducing discoveries from gene expression analysis. https://t.co/ilVZoItFws

[ML-Bioinfo-CEITEC @MCeitec]

#GunmaUniversity #UniversityofToyama @KedokteranUI developed #DL programs to extract radiosensitivity data from literature.Programs #1–3 screen papers with data obtained by CAs.Program #4 extracts CA-derived SF2 data from semi-logarithmic survival curves.https://t.co/ltif5BlGNi

[ML-Bioinfo-CEITEC @MCeitec]

@UniHeidelberg @dzhk_germany @Saar_Uni @SiemensHealth @Stanford developed #NN model incorporating 34 validated ACS miRNAs.The additional #ML models achieved an accuracy of 0.96 (95% CI 0.96–0.97), sensitivity of 0.95, specificity of 0.96, and AUC of 0.99.
https://t.co/FJIg5ddSzK

[ML-Bioinfo-CEITEC @MCeitec]

Don’t miss this great read in Towards Data Science written by Renu Khandelwal titled “Inferences from a TF Lite model — Transfer Learning on a Pre-trained Model”
@TDataScience https://t.co/RXAR0Ksxmy

[ML-Bioinfo-CEITEC @MCeitec]

Congratulations to Vojtěch and the whole team.

[CEITEC @CEITEC_Brno]

Scientists, doctors, and informaticians in Brno develop unique #BioinformaticsSoftware. It will serve as a more sophisticated treatment planning tool for paediatric #oncology patients. A team of scientists is led by Vojtech Bystry, @CEITEC_Brno @muni_cz.
https://t.co/05ZRsvjMLm

[ML-Bioinfo-CEITEC @MCeitec]

@uarizona presented Artificial Neurogenesis (ANG) algorithm, that grows rather than prunes the network and enables neural networks to be trained and executed in low SWaP embedded hardware.
https://t.co/sC24KnI15G

[ML-Bioinfo-CEITEC @MCeitec]

@EdinburghUni proposed a training set synthesis technique, called Dataset Condensation, that learns to produce a small set of informative samples for training #DNN at small computational cost on the original data while achieving comparable results. https://t.co/nX2AYLgojn

[ML-Bioinfo-CEITEC @MCeitec]

National Research Council of Italy @NOVAunl
applied two #DL classifiers and three #ML models to two different miRNA-mRNA datasets, of predictions from 3 tools: TargetScan, miRanda, and RNAhybrid to compare the two approaches in Bioinformatics: https://t.co/yjnnUk6ISR

[ML-Bioinfo-CEITEC @MCeitec]

The future of machine learning by @jeremiecharris in @TDataScience https://t.co/5JSeXnLeYQ

[ML-Bioinfo-CEITEC @MCeitec]

@HHMIJanelia @embl present ilastik, an interactive tool that brings #ML-based (bio)image analysis to users without substantial computational expertise. It contains pre-defined workflows for image segmentation, object classification, counting and tracking.
https://t.co/DACvNVHBmA