KNIME for reproducible cross-domain analysis of life science data

Fillbrunn, Alexander; Dietz, Christian; Pfeuffer, Julianus; Rahn, René; Landrum, Gregory A.; Berthold, Michael R.

doi:10.1016/j.jbiotec.2017.07.028

Cited by 160 publications

(106 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… The use of well documented and amenable workflow management platforms like KNIME facilitate the construction of consistent, reproducible, and transferable protocols . The workflows can be transferred between, for example, workstations, users, and sites, and can be re‐run: i) as is, for example, when large data transfer is not feasible, or when new database versions are released; ii) with different configurations of the nodes, for example, changing ligand activity cut‐offs (Figure ), input ligands (Figures , , ), protein targets (Figure ); iii) with additional/modified nodes to obtain complementary information, for example, including annotations from other databases, further analyzing results, or performing machine learning on the obtained data.…”

Section: Discussionmentioning

confidence: 99%

“…There is a need for eScience technologies to process the large volumes of rapidly generated, heterogeneous protein–ligand interaction data into computational models that enable the design of efficacious and safe medicines . The ChEMBL database (version 23), for example, contains over 14 million data entries on 11 500 protein targets, of which 4600 human, covering 1.7 million unique compounds .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3D‐e‐Chem: Structural Cheminformatics Workflows for Computer‐Aided Drug Discovery

et al. 2018

View full text Add to dashboard Cite

AbstracteScience technologies are needed to process the information available in many heterogeneous types of protein–ligand interaction data and to capture these data into models that enable the design of efficacious and safe medicines. Here we present scientific KNIME tools and workflows that enable the integration of chemical, pharmacological, and structural information for: i) structure‐based bioactivity data mapping, ii) structure‐based identification of scaffold replacement strategies for ligand design, iii) ligand‐based target prediction, iv) protein sequence‐based binding site identification and ligand repurposing, and v) structure‐based pharmacophore comparison for ligand repurposing across protein families. The modular setup of the workflows and the use of well‐established standards allows the re‐use of these protocols and facilitates the design of customized computer‐aided drug discovery workflows.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

3D‐e‐Chem: Structural Cheminformatics Workflows for Computer‐Aided Drug Discovery

et al. 2018

View full text Add to dashboard Cite

show abstract

“…First, preliminary labels were automatically generated using either classical image processing techniques, or existing, non-trained nuclear segmentation deep learning models. In particular, since nuclei of MCF10A cells displayed high contrast and good separation, ground truth labels (instance segmentation masks) for these cells were generated using a simple nuclear segmentation pipeline in KNIME (19,20), which included thresholding, gaps filling, connected components, and manual separation of a few adjacent nuclei. A set of Python scripts was used to convert these preliminary nuclear labels images into a format compatible with the interactive, web-based image editing tool Supervisely (21).…”

Section: Ground Truth Labels Generationmentioning

confidence: 99%

A Deep Learning Pipeline for Nucleus Segmentation

Zaki

Gudla

Lee

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep learning is rapidly becoming the technique of choice for automated segmentation of nuclei in biological image analysis workflows. In order to improve and understand the training parameters that drive the performance of deep learning models trained on small, custom annotated image datasets, we have designed a computational pipeline to systematically test different nuclear segmentation model architectures and model training strategies. Using this approach, we demonstrate that transfer learning and tuning of training parameters, such as the training image dataset composition, size and pre-processing, can lead to robust nuclear segmentation models, which match, and often exceed, the performance of existing, state-of-the-art deep learning models pre-trained on large image datasets. Our work provides computational tools and a practical framework for the improvement of deep learning-based biological image segmentation using small annotated image datasets.

show abstract

“…Moreover, a RESTful service has been implemented, so that Selenzyme can accept multiple queries from any other web-based application. As an example of the application of the REST service, a KNIME node (O'Hagan and Kell, 2015;Fillbrunn et al, 2017) is available so that the reaction query can be generated from chemoinformatics operations within a workflow (examples available at http://www.myexperiment.org/packs/734, see supplementary information). Similarly, the resulting output tables containing the sequences can easily be processed downstream in the workflow or sent to other sequence analysis services such as Galaxy.…”

Section: Web Server and Restful Servicementioning

confidence: 99%

Selenzyme: Enzyme selection tool for pathway design

Carbonell

Wong

Swainston

et al. 2017

Preprint

View full text Add to dashboard Cite

Synthetic biology applies the principles of engineering to biology in order to create biological functionalities not seen before in nature. One of the most exciting applications of synthetic biology is the design of new organisms with the ability to produce valuable chemicals including pharmaceuticals and biomaterials in a greener; sustainable fashion. Selecting the right enzymes to catalyze each reaction step in order to produce a desired target compound is, however, not trivial. Here, we present Selenzyme, a free online enzyme selection tool for metabolic pathway design. The user is guided through several decision steps in order to shortlist the best candidates for a given pathway step. The tool graphically presents key information about enzymes based on existing databases and tools such as: similarity of sequences and of catalyzed reactions; phylogenetic distance between source organism and intended host species; multiple alignment highlighting conserved regions, predicted catalytic site, and active regions; and relevant properties such as predicted solubility and transmembrane regions. Selenzyme provides bespoke sequence selection for automated workflows in biofoundries. The tool is integrated as part of the pathway design stage into the design-build-

show abstract

KNIME for reproducible cross-domain analysis of life science data

Cited by 160 publications

References 37 publications

3D‐e‐Chem: Structural Cheminformatics Workflows for Computer‐Aided Drug Discovery

3D‐e‐Chem: Structural Cheminformatics Workflows for Computer‐Aided Drug Discovery

A Deep Learning Pipeline for Nucleus Segmentation

Selenzyme: Enzyme selection tool for pathway design

Contact Info

Product

Resources

About