Related Experiment Video
Updated: Jun 25, 2026
Cell Based Assays of SINEUP Non-coding RNAs That Can Specifically Enhance mRNA Translation
Published on: February 1, 2019
Computational discovery of sense-antisense transcription in the human and mouse genomes
Jay Shendure1, George M Church
1Department of Genetics, Harvard Medical School, Boston, MA 02115, USA.
Researchers used computer analysis of public genetic databases to identify hundreds of new instances where genes are transcribed in opposite directions, potentially creating double-stranded RNA structures that regulate gene activity.
Area of Science:
- Genomics and bioinformatics research within sense-antisense transcription studies
- Computational biology and molecular genetics
Background:
No prior work had fully mapped the extent of bidirectional gene expression across mammalian genomes. That uncertainty drove interest in how these overlapping transcripts influence cellular regulation. It was already known that double-stranded RNA duplexes participate in diverse processes like genomic imprinting and RNA interference. Prior research has shown these structures also affect translational control and alternative splicing. This gap motivated a systematic search for additional genomic regions exhibiting this orientation. Scientists previously identified limited examples of such phenomena in various species. However, the total prevalence of these overlapping units remained largely uncharacterized. This study addresses that void by mining large-scale sequence databases for novel candidates.
Purpose Of The Study:
The aim of this study is to discover and characterize sense-antisense transcription across human and mouse genomes. Researchers sought to address the lack of comprehensive data regarding bidirectionally transcribed genomic regions. This investigation was motivated by the increasing recognition of double-stranded RNA in various regulatory phenomena. The team intended to expand the catalog of known overlapping transcriptional units using large-scale data mining. They aimed to determine if these overlaps correlate with evolutionary conservation in untranslated regions. By applying bioinformatics techniques, the authors hoped to provide a more accurate estimate of the prevalence of these structures. The study addresses the need for a systematic approach to identify novel regulatory elements. Ultimately, the researchers intended to test the hypothesis that these overlaps explain specific patterns of genomic conservation.
Main Methods:
Review approach involved mining public expressed sequence tag repositories for bidirectional transcriptional activity. The team implemented a bioinformatics pipeline to scan human and mouse genetic information systematically. They filtered these datasets to isolate overlapping units oriented in opposite directions. Researchers then selected a subset of these predictions for laboratory verification. This validation step utilized orientation-specific reverse transcription polymerase chain reaction to confirm the existence of the transcripts. The investigators calculated the specificity of their computational model based on these experimental results. They compared the identified overlaps against known genomic maps to assess evolutionary conservation. This multi-step strategy ensured that the predicted candidates were both statistically significant and biologically plausible.
Main Results:
Key findings from the literature reveal the identification of over 217 novel candidate overlapping transcriptional units. The methodology achieved a specificity of 84% or greater during experimental validation. The study brings the total number of predicted and validated examples to over 300 across the two species. Many of these overlaps occur within the 5' or 3' untranslated regions of transcripts. The researchers observed a strong correlation between these overlaps and patterns of mouse-human genomic conservation. These results suggest that bidirectional transcription is a widespread phenomenon in mammalian genomes. The data indicate that most of the identified transcriptional units were previously unknown. These findings provide a comprehensive catalog of potential double-stranded RNA forming regions.
Conclusions:
The authors propose that their computational approach successfully identifies hundreds of novel bidirectionally transcribed genomic regions. Synthesis and implications suggest that these overlapping units contribute to the observed patterns of evolutionary conservation. Researchers indicate that a significant portion of conserved non-coding sequences may actually represent functional transcriptional units. The data support the hypothesis that sense-antisense interactions play a role in complex gene regulation. Evidence points toward these structures being more common than previously documented in mammalian genomes. The study demonstrates that bioinformatics tools can effectively predict these interactions with high specificity. These findings provide a framework for future investigations into the functional consequences of double-stranded RNA. The authors conclude that their work expands the known landscape of mammalian gene expression regulation.
Frequently Asked Questions
The researchers propose that overlapping transcripts form double-stranded RNA duplexes, which potentially regulate gene activity. This mechanism is linked to processes like RNA interference and genomic imprinting, distinguishing it from simple protein-coding gene expression.
The team utilized expressed sequence tag databases to identify candidate regions. This computational approach allowed them to screen large genomic datasets, whereas traditional laboratory methods would be too slow to identify over 217 novel transcriptional units.
Orientation-specific reverse transcription polymerase chain reaction was necessary to confirm the accuracy of the predictions. This technique distinguishes between the two strands, providing an 84% specificity rate, which is higher than standard non-directional sequencing methods.
Expressed sequence tag data served as the primary input for the bioinformatics pipeline. These sequences provide a snapshot of transcribed regions, enabling the detection of bidirectional overlaps that might be missed by analyzing only genomic DNA sequences.
The researchers measured the correlation between sense-antisense overlaps and mouse-human conservation. They observed that many overlaps in untranslated regions align with conserved genomic patterns, suggesting these regions maintain functional importance across different mammalian species.
The authors propose that a subset of conserved sequences in untranslated regions might be explained by the presence of an overlapping transcriptional unit. This implies that some evolutionary conservation is driven by regulatory RNA rather than protein-coding requirements.
