UH Biocomputation Group - genericshttp://biocomputation.herts.ac.uk/2023-02-08T13:28:44+00:00miRe2e: A Full End-to-End Deep Model Based on Transformers for Prediction of Pre-miRNAs2023-02-08T13:28:44+00:002023-02-08T13:28:44+00:00Manal Helaltag:biocomputation.herts.ac.uk,2023-02-08:/2023/02/08/mire2e-a-full-end-to-end-deep-model-based-on-transformers-for-prediction-of-pre-mirnas.html<p class="first last">Manal Helal's Journal Club session where she will talk about a paper "miRe2e: A Full End-to-End Deep Model Based on Transformers for Prediction of Pre-miRNAs"</p>
<p>This week on Journal Club session Manal Helal will talk about a paper "miRe2e: A Full End-to-End Deep Model Based on Transformers for Prediction of Pre-miRNAs".</p>
<hr class="docutils" />
<p><strong>Motivation</strong></p>
<p>MicroRNAs (miRNAs) are small RNA sequences with key roles in the regulation of
gene expression at post-transcriptional level in different species. Accurate
prediction of novel miRNAs is needed due to their importance in many biological
processes and their associations with complicated diseases in humans. Many
machine learning approaches were proposed in the last decade for this purpose,
but requiring handcrafted features extraction to identify possible de novo
miRNAs. More recently, the emergence of deep learning (DL) has allowed the
automatic feature extraction, learning relevant representations by themselves.
However, the state-of-art deep models require complex pre-processing of the
input sequences and prediction of their secondary structure to reach an
acceptable performance.</p>
<p><strong>Results</strong></p>
<p>In this work, we present miRe2e, the first full end-to-end DL model for
pre-miRNA prediction. This model is based on Transformers, a neural
architecture that uses attention mechanisms to infer global dependencies
between inputs and outputs. It is capable of receiving the raw genome-wide data
as input, without any pre-processing nor feature engineering. After a training
stage with known pre-miRNAs, hairpin and non-harpin sequences, it can identify
all the pre-miRNA sequences within a genome. The model has been validated
through several experimental setups using the human genome, and it was compared
with state-of-the-art algorithms obtaining 10 times better performance.</p>
<div class="line-block">
<div class="line"><br /></div>
</div>
<p>Papers:</p>
<ul class="simple">
<li>J. Raad, L. Bugnon, D. Milone, G. Stegmayer, <a class="reference external" href="https://doi.org/10.1093/bioinformatics/btab823">"miRe2e: A Full End-to-End Deep
Model Based on Transformers for Prediction of Pre-miRNAs"</a>, 2022, Bioinformatics,
38, 1191--1197</li>
</ul>
<p><strong>Date:</strong> 2023/02/10 <br />
<strong>Time:</strong> 14:00 <br />
<strong>Location</strong>: online</p>