See the published version of this preprint at BioData Mining.
This is a preprint, a preliminary version of a manuscript that has not completed peer review at a journal. Research Square does not conduct peer review prior to posting preprints. The posting of a preprint on this server should not be interpreted as an endorsement of its validity or suitability for dissemination as established information or for guiding clinical practice.
Learn more about In Review
A new preprint design is coming!
We have improved readability, clarity, accessibility, and opportunities for engagement.
Research
Shabir Moosa
Sidra Medical and Research Center
Corresponding Author
ORCiD: https://orcid.org/0000-0003-2645-2497
License:
This work is licensed under a CC BY 4.0 License. Read Full License
Background: The data explosion caused by unprecedented advancements in the field of genomics is constantly challenging the conventional methods used in the interpretation of the human genome. The demand for robust algorithms over the recent years has brought huge success in the field of Deep Learning (DL) in solving many difficult tasks in image, speech and natural language processing by automating the manual process of architecture design. This has been fueled through the development of new DL architectures. Yet genomics possesses unique challenges as we expect DL to provide a super human intelligence that easily interprets a human genome.
Methods: We proposed a new model, DASSI, by adapting a differential architecture search method and applying it to the Splice Site (SS) recognition task on DNA sequences to discover new high-performance convolutional architectures in an automated manner. We evaluated the discovered model against state-of-the-art tools to classify true and false SS in Homo sapiens (Human), Arabidopsis thaliana (Plant), Caenorhabditis elegans (Worm) and Drosophila melanogaster (Fly).
Results: Our experimental evaluation demonstrated that the discovered architecture outperformed baseline models and fixed architectures and showed competitive results against state-of-the-art models used in classification of splice sites. The proposed model - DASSI has a compact architecture and showed very good results on a transfer learning task. The benchmarking experiments of execution time and precision on architecture search and evaluation process showed better performance on recently available GPUs making it feasible to adopt such architecture search based methods for large datasets.
Conclusions: We applied differential architecture search mechanism to perform SS classification on raw DNA sequences, and discovered new convolutional models with low number of tunable parameters and competitive performance compared with manually engineered architectures. The results have shown a potential of using this automated architecture search mechanism for solving various problems in the field of genomics.
Keywords
Deep Learning, Splice Site, Genomics, Neural Architecture Search, Convolutional Neural Networks
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Loading...
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
Peer Review Timeline
CURRENT STATUS: Published
View the published article at BioData Mining.
No community comments so far
Editorial decision: Accept
On 03 Jan, 2021
Reviewer #2 agreed
On 15 Nov, 2020
Reviewer #1 agreed
On 15 Nov, 2020
Review #1 received
Received 15 Nov, 2020
Review #2 received
Received 15 Nov, 2020
Editor assigned
On 14 Nov, 2020
Reviewers invited
Invitations sent on 14 Nov, 2020
Submission checks complete
On 14 Nov, 2020
Editor invited
On 14 Nov, 2020
Version 2
Posted 19 Oct, 2020
No comments provided
Review #1 received
Received 17 Oct, 2020
Review #3 received
Received 17 Oct, 2020
Editorial decision: Minor revision
On 17 Oct, 2020
Reviewer #3 agreed
On 14 Oct, 2020
Reviewer #2 agreed
On 12 Oct, 2020
Review #2 received
Received 12 Oct, 2020
Editor assigned
On 11 Oct, 2020
Reviewers invited
Invitations sent on 11 Oct, 2020
Reviewer #1 agreed
On 11 Oct, 2020
Submission checks complete
On 10 Oct, 2020
Editor invited
On 10 Oct, 2020
No comments provided
Review #3 received
Received 03 Aug, 2020
Editorial decision: Major revision
On 03 Aug, 2020
Review #2 received
Received 01 Aug, 2020
Reviewer #3 agreed
On 21 Jul, 2020
Reviewer #2 agreed
On 19 Jul, 2020
Review #1 received
Received 18 Jul, 2020
Reviewer #1 agreed
On 17 Jul, 2020
Reviewers invited
Invitations sent on 16 Jul, 2020
Editor assigned
On 15 Jul, 2020
Submission checks complete
On 14 Jul, 2020
Editor invited
On 14 Jul, 2020
First submitted
On 01 Jul, 2020
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
Subject Areas
Bioinformatics
Learn more about our company.
A new preprint design is coming!
We have improved readability, clarity, accessibility, and opportunities for engagement.
See the published version of this preprint at BioData Mining.
This is a preprint, a preliminary version of a manuscript that has not completed peer review at a journal. Research Square does not conduct peer review prior to posting preprints. The posting of a preprint on this server should not be interpreted as an endorsement of its validity or suitability for dissemination as established information or for guiding clinical practice.
Learn more about In Review
Research
Shabir Moosa
Sidra Medical and Research Center
Corresponding Author
ORCiD: https://orcid.org/0000-0003-2645-2497
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
Peer Review Timeline
CURRENT STATUS: Published
View the published article at BioData Mining.
No community comments so far
Editorial decision: Accept
On 03 Jan, 2021
Reviewer #2 agreed
On 15 Nov, 2020
Reviewer #1 agreed
On 15 Nov, 2020
Review #1 received
Received 15 Nov, 2020
Review #2 received
Received 15 Nov, 2020
Editor assigned
On 14 Nov, 2020
Reviewers invited
Invitations sent on 14 Nov, 2020
Submission checks complete
On 14 Nov, 2020
Editor invited
On 14 Nov, 2020
Version 2
Posted 19 Oct, 2020
No comments provided
Review #1 received
Received 17 Oct, 2020
Review #3 received
Received 17 Oct, 2020
Editorial decision: Minor revision
On 17 Oct, 2020
Reviewer #3 agreed
On 14 Oct, 2020
Reviewer #2 agreed
On 12 Oct, 2020
Review #2 received
Received 12 Oct, 2020
Editor assigned
On 11 Oct, 2020
Reviewers invited
Invitations sent on 11 Oct, 2020
Reviewer #1 agreed
On 11 Oct, 2020
Submission checks complete
On 10 Oct, 2020
Editor invited
On 10 Oct, 2020
No comments provided
Review #3 received
Received 03 Aug, 2020
Editorial decision: Major revision
On 03 Aug, 2020
Review #2 received
Received 01 Aug, 2020
Reviewer #3 agreed
On 21 Jul, 2020
Reviewer #2 agreed
On 19 Jul, 2020
Review #1 received
Received 18 Jul, 2020
Reviewer #1 agreed
On 17 Jul, 2020
Reviewers invited
Invitations sent on 16 Jul, 2020
Editor assigned
On 15 Jul, 2020
Submission checks complete
On 14 Jul, 2020
Editor invited
On 14 Jul, 2020
First submitted
On 01 Jul, 2020
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
Subject Areas
Bioinformatics
License:
This work is licensed under a CC BY 4.0 License. Read Full License
View the published article at BioData Mining.
Background: The data explosion caused by unprecedented advancements in the field of genomics is constantly challenging the conventional methods used in the interpretation of the human genome. The demand for robust algorithms over the recent years has brought huge success in the field of Deep Learning (DL) in solving many difficult tasks in image, speech and natural language processing by automating the manual process of architecture design. This has been fueled through the development of new DL architectures. Yet genomics possesses unique challenges as we expect DL to provide a super human intelligence that easily interprets a human genome.
Methods: We proposed a new model, DASSI, by adapting a differential architecture search method and applying it to the Splice Site (SS) recognition task on DNA sequences to discover new high-performance convolutional architectures in an automated manner. We evaluated the discovered model against state-of-the-art tools to classify true and false SS in Homo sapiens (Human), Arabidopsis thaliana (Plant), Caenorhabditis elegans (Worm) and Drosophila melanogaster (Fly).
Results: Our experimental evaluation demonstrated that the discovered architecture outperformed baseline models and fixed architectures and showed competitive results against state-of-the-art models used in classification of splice sites. The proposed model - DASSI has a compact architecture and showed very good results on a transfer learning task. The benchmarking experiments of execution time and precision on architecture search and evaluation process showed better performance on recently available GPUs making it feasible to adopt such architecture search based methods for large datasets.
Conclusions: We applied differential architecture search mechanism to perform SS classification on raw DNA sequences, and discovered new convolutional models with low number of tunable parameters and competitive performance compared with manually engineered architectures. The results have shown a potential of using this automated architecture search mechanism for solving various problems in the field of genomics.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Loading...