Skip to main content
Premium Trial:

Request an Annual Quote

Tools & Techniques: RNAi Screen Data Visualization, microRNA Classification, and More


Given the abundance and complexity of data resulting from large-scale RNAi screens, a team led by an investigator from the Technical University of Dresden reported on the development of an interactive visualization methodology for the analysis of high-content screening datasets.

According to a paper in BioMed Research International, the high dimensionality of these datasets often prevents users from recognizing important patterns in them, while understanding the importance of such patterns represents “an even more difficult problem.”

To address such issues, the researchers developed a method called 1Click1View, which involves providing data in one visual frame that “allows users to gain insight into the data and generate hypotheses by directly interacting with image data.”

Through the approach, users are “directly involved" in the imaging processing results “to combine the flexibility, creativity, and general knowledge of the scientist with the enormous amount of numerical rows connected to image files,” the authors wrote.

In 1Click1View, original image data, their image processing results, and metadata are joined together and are available for filtering and clustering in an interactive view. The method offers an “effective and efficient method for interactive image data exploration, detection of systematic errors, and quality control,” the paper states.

A group of researchers from Kyushu University has published a new method for enabling the uptake of dsRNA via soaking in Sf9 cells, a clonal isolate of Spodoptera frugiperda cells that are widely used for the production of recombinant proteins using baculovirus.

Passive dsRNA uptake occurs in Caenorhabditis elegans due to the presence of the dsRNA transport protein SID-1, the investigators wrote in Applied Microbiology and Biotechnology. In insects, however, triggering RNAi requires the transfection of dsRNA, which is often expensive and cytotoxic.

Looking to overcome this hurdle, the research team ectopically expressed C. elegans SID-1 in Sf9 cells, which endowed them with the capacity for soaking RNAi. This allowed the scientists to modify target proteins of a baculovirus expression vector system in both quantities and post-transational modifications.

“There can be no doubt that the wide application of Sf9 cells in gene functional research will provide new insights into many canonical pathways among insect cell lines,” they wrote. Their approach has resulted in a “low-cost and high-efficiency RNAi system … useful for high-throughput gene functional analysis and mass production of recombinant protein.”

Despite the potential biological importance of interactions between microRNA and transcription factors, studying the regulatory mechanism that involve the two remains a challenge.

As such, a group of Australian researchers has proposed a framework to infer gene regulatory networks involving both transcription factors and miRNAs from multiple sources of data, including gene expression data, target information, and sample categories, according to a paper in BMC Bioinformatics.

The framework involves preparing the input for the “network structure learning, including collecting target information for [transcription factors] and miRNAs, normalizing expression data, and analyzing differentially expressed genes,” the scientists wrote.

Next, target information is transformed into five types of network substructures and datasets are split according to sample conditions, resulting in an “integrated global network” made of miRNAs, transcription factors, and mRNAs.

Lastly, the global network is searched for subgraphs that show the interplay between miRNAs and transcription factors, and network motifs that involve at least two regulators.

A team of Polish investigators has published details of a new machine learning microRNA classification tool called HuntMi, which they say offers better performance than other tools and can be used in experiments in a range of species.

“Machine learning techniques are known to be a powerful way of distinguishing microRNA hairpins from pseudo hairpins and have been applied in a number of recognized miRNA search tools,” they wrote in BMC Bioinformatics. “However, many current methods based on machine learning suffer from some drawbacks, including not addressing the class imbalance problem properly.”

HuntMi employs a unique strategy of thresholding score functions produced by traditional classifiers, as well as new features for data representation that further improve classification performance, according to the paper.

“The method was tested on large and strongly imbalanced datasets using stratified 10-fold cross-validation procedure,” the researchers noted. “Classification performance was further verified on miRNAs newly introduced in [the] latest builds of miRBase. As a result, HuntMi clearly outperforms state-of-the-art miRNA hairpin classification tools … without compromising the training time.”

The tool comes optimized for humans and Arabidopsis, as well as various other animals, plants, and viruses, and there is a possibility to train a model on any dataset and use it in classification analysis.