DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks
Development of automated analysis tools for “single ion channel” recording is hampered by the lack of available training data. For machine learning based tools, very large training sets are necessary with sample-by-sample point labelled data (e.g., 1 sample point every 100microsecond). In an experim...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9089889/?tool=EBI |
| _version_ | 1811247704239505408 |
|---|---|
| author | Sam T. M. Ball Numan Celik Elaheh Sayari Lina Abdul Kadir Fiona O’Brien Richard Barrett-Jolley |
| author_facet | Sam T. M. Ball Numan Celik Elaheh Sayari Lina Abdul Kadir Fiona O’Brien Richard Barrett-Jolley |
| author_sort | Sam T. M. Ball |
| collection | DOAJ |
| description | Development of automated analysis tools for “single ion channel” recording is hampered by the lack of available training data. For machine learning based tools, very large training sets are necessary with sample-by-sample point labelled data (e.g., 1 sample point every 100microsecond). In an experimental context, such data are labelled with human supervision, and whilst this is feasible for simple experimental analysis, it is infeasible to generate the enormous datasets that would be necessary for a big data approach using hand crafting. In this work we aimed to develop methods to generate simulated ion channel data that is free from assumptions and prior knowledge of noise and underlying hidden Markov models. We successfully leverage generative adversarial networks (GANs) to build an end-to-end pipeline for generating an unlimited amount of labelled training data from a small, annotated ion channel “seed” record, and this needs no prior knowledge of theoretical dynamical ion channel properties. Our method utilises 2D CNNs to maintain the synchronised temporal relationship between the raw and idealised record. We demonstrate the applicability of the method with 5 different data sources and show authenticity with t-SNE and UMAP projection comparisons between real and synthetic data. The model would be easily extendable to other time series data requiring parallel labelling, such as labelled ECG signals or raw nanopore sequencing data. |
| first_indexed | 2024-04-12T15:14:19Z |
| format | Article |
| id | doaj.art-a9ee232208dc44bf9e9c91368f2ec904 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-12T15:14:19Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-a9ee232208dc44bf9e9c91368f2ec9042022-12-22T03:27:39ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-01175DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networksSam T. M. BallNuman CelikElaheh SayariLina Abdul KadirFiona O’BrienRichard Barrett-JolleyDevelopment of automated analysis tools for “single ion channel” recording is hampered by the lack of available training data. For machine learning based tools, very large training sets are necessary with sample-by-sample point labelled data (e.g., 1 sample point every 100microsecond). In an experimental context, such data are labelled with human supervision, and whilst this is feasible for simple experimental analysis, it is infeasible to generate the enormous datasets that would be necessary for a big data approach using hand crafting. In this work we aimed to develop methods to generate simulated ion channel data that is free from assumptions and prior knowledge of noise and underlying hidden Markov models. We successfully leverage generative adversarial networks (GANs) to build an end-to-end pipeline for generating an unlimited amount of labelled training data from a small, annotated ion channel “seed” record, and this needs no prior knowledge of theoretical dynamical ion channel properties. Our method utilises 2D CNNs to maintain the synchronised temporal relationship between the raw and idealised record. We demonstrate the applicability of the method with 5 different data sources and show authenticity with t-SNE and UMAP projection comparisons between real and synthetic data. The model would be easily extendable to other time series data requiring parallel labelling, such as labelled ECG signals or raw nanopore sequencing data.https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9089889/?tool=EBI |
| spellingShingle | Sam T. M. Ball Numan Celik Elaheh Sayari Lina Abdul Kadir Fiona O’Brien Richard Barrett-Jolley DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks PLoS ONE |
| title | DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks |
| title_full | DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks |
| title_fullStr | DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks |
| title_full_unstemmed | DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks |
| title_short | DeepGANnel: Synthesis of fully annotated single molecule patch-clamp data using generative adversarial networks |
| title_sort | deepgannel synthesis of fully annotated single molecule patch clamp data using generative adversarial networks |
| url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9089889/?tool=EBI |
| work_keys_str_mv | AT samtmball deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks AT numancelik deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks AT elahehsayari deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks AT linaabdulkadir deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks AT fionaobrien deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks AT richardbarrettjolley deepgannelsynthesisoffullyannotatedsinglemoleculepatchclampdatausinggenerativeadversarialnetworks |