Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge
Abstract Background Proteomic measurements, which closely reflect phenotypes, provide insights into gene expression regulations and mechanisms underlying altered phenotypes. Further, integration of data on proteome and transcriptome levels can validate gene signatures associated with a phenotype. Ho...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2019-12-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12859-019-3253-z |
| _version_ | 1818601231729295360 |
|---|---|
| author | Tara Eicher Andrew Patt Esko Kautto Raghu Machiraju Ewy Mathé Yan Zhang |
| author_facet | Tara Eicher Andrew Patt Esko Kautto Raghu Machiraju Ewy Mathé Yan Zhang |
| author_sort | Tara Eicher |
| collection | DOAJ |
| description | Abstract Background Proteomic measurements, which closely reflect phenotypes, provide insights into gene expression regulations and mechanisms underlying altered phenotypes. Further, integration of data on proteome and transcriptome levels can validate gene signatures associated with a phenotype. However, proteomic data is not as abundant as genomic data, and it is thus beneficial to use genomic features to predict protein abundances when matching proteomic samples or measurements within samples are lacking. Results We evaluate and compare four data-driven models for prediction of proteomic data from mRNA measured in breast and ovarian cancers using the 2017 DREAM Proteogenomics Challenge data. Our results show that Bayesian network, random forests, LASSO, and fuzzy logic approaches can predict protein abundance levels with median ground truth-predicted correlation values between 0.2 and 0.5. However, the most accurately predicted proteins differ considerably between approaches. Conclusions In addition to benchmarking aforementioned machine learning approaches for predicting protein levels from transcript levels, we discuss challenges and potential solutions in state-of-the-art proteogenomic analyses. |
| first_indexed | 2024-12-16T12:48:06Z |
| format | Article |
| id | doaj.art-08bc8c48c42543ad81b05381facb78c2 |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-12-16T12:48:06Z |
| publishDate | 2019-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-08bc8c48c42543ad81b05381facb78c22022-12-21T22:31:14ZengBMCBMC Bioinformatics1471-21052019-12-0120S2411610.1186/s12859-019-3253-zChallenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challengeTara Eicher0Andrew Patt1Esko Kautto2Raghu Machiraju3Ewy Mathé4Yan Zhang5Department of Computer Science and Engineering, The Ohio State UniversityDepartment of Biomedical Informatics, College of Medicine, The Ohio State UniversityDepartment of Biomedical Informatics, College of Medicine, The Ohio State UniversityDepartment of Computer Science and Engineering, The Ohio State UniversityDepartment of Biomedical Informatics, College of Medicine, The Ohio State UniversityDepartment of Biomedical Informatics, College of Medicine, The Ohio State UniversityAbstract Background Proteomic measurements, which closely reflect phenotypes, provide insights into gene expression regulations and mechanisms underlying altered phenotypes. Further, integration of data on proteome and transcriptome levels can validate gene signatures associated with a phenotype. However, proteomic data is not as abundant as genomic data, and it is thus beneficial to use genomic features to predict protein abundances when matching proteomic samples or measurements within samples are lacking. Results We evaluate and compare four data-driven models for prediction of proteomic data from mRNA measured in breast and ovarian cancers using the 2017 DREAM Proteogenomics Challenge data. Our results show that Bayesian network, random forests, LASSO, and fuzzy logic approaches can predict protein abundance levels with median ground truth-predicted correlation values between 0.2 and 0.5. However, the most accurately predicted proteins differ considerably between approaches. Conclusions In addition to benchmarking aforementioned machine learning approaches for predicting protein levels from transcript levels, we discuss challenges and potential solutions in state-of-the-art proteogenomic analyses.https://doi.org/10.1186/s12859-019-3253-zProteogenomicsmRNARandom forestsFuzzy logicBayesian networks |
| spellingShingle | Tara Eicher Andrew Patt Esko Kautto Raghu Machiraju Ewy Mathé Yan Zhang Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge BMC Bioinformatics Proteogenomics mRNA Random forests Fuzzy logic Bayesian networks |
| title | Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge |
| title_full | Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge |
| title_fullStr | Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge |
| title_full_unstemmed | Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge |
| title_short | Challenges in proteogenomics: a comparison of analysis methods with the case study of the DREAM proteogenomics sub-challenge |
| title_sort | challenges in proteogenomics a comparison of analysis methods with the case study of the dream proteogenomics sub challenge |
| topic | Proteogenomics mRNA Random forests Fuzzy logic Bayesian networks |
| url | https://doi.org/10.1186/s12859-019-3253-z |
| work_keys_str_mv | AT taraeicher challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge AT andrewpatt challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge AT eskokautto challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge AT raghumachiraju challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge AT ewymathe challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge AT yanzhang challengesinproteogenomicsacomparisonofanalysismethodswiththecasestudyofthedreamproteogenomicssubchallenge |