The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines
Background: Gene expression signatures are clusters of genes discriminating different statuses of the cells and their definition is critical for understanding the molecular bases of diseases. The identification of a gene signature is complicated by the high dimensional nature of the data and by the...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
BioMed Central
2010
|
| Online Access: | http://hdl.handle.net/1721.1/52504 https://orcid.org/0000-0001-6376-4786 |
| _version_ | 1826215351179804672 |
|---|---|
| author | Varesio, Luigi Verri, Alessandro Mosci, Sofia Barla, Annalisa Fardin, Paolo Rosasco, Lorenzo Andrea |
| author2 | Massachusetts Institute of Technology. Center for Biological & Computational Learning |
| author_facet | Massachusetts Institute of Technology. Center for Biological & Computational Learning Varesio, Luigi Verri, Alessandro Mosci, Sofia Barla, Annalisa Fardin, Paolo Rosasco, Lorenzo Andrea |
| author_sort | Varesio, Luigi |
| collection | MIT |
| description | Background: Gene expression signatures are clusters of genes discriminating different statuses of the cells and their definition is critical for understanding the molecular bases of diseases. The identification of a gene signature is complicated by the high dimensional nature of the data and by the genetic heterogeneity of the responding cells. The l[subscript 1]-l[subscript 2] regularization is an embedded feature selection technique that fulfills all the desirable properties of a variable selection algorithm and has the potential to generate a specific signature even in biologically complex settings. We studied the application of this algorithm to detect the signature characterizing the transcriptional response of neuroblastoma tumor cell lines to hypoxia, a condition of low oxygen tension that occurs in the tumor microenvironment.
Results: We determined the gene expression profile of 9 neuroblastoma cell lines cultured under normoxic and hypoxic conditions. We studied a heterogeneous set of neuroblastoma cell lines to mimic the in vivo situation and to test the robustness and validity of the l[subscript 1]-l[subscript 2] regularization with double optimization. Analysis by hierarchical, spectral, and k-means clustering or supervised approach based on t-test analysis divided the cell lines on the bases of genetic differences. However, the disturbance of this strong transcriptional response completely masked the detection of the more subtle response to hypoxia. Different results were obtained when we applied the l[subscript 1]-l[subscript 2] regularization framework. The algorithm distinguished the normoxic and hypoxic statuses defining signatures comprising 3 to 38 probesets, with a leave-one-out error of 17%. A consensus hypoxia signature was established setting the frequency score at 50% and the correlation parameter ε equal to 100. This signature is composed by 11 probesets representing 8 well characterized genes known to be modulated by hypoxia.
Conclusion: We demonstrate that l[subscript 1]-l[subscript 2] regularization outperforms more conventional approaches allowing the identification and definition of a gene expression signature under complex experimental conditions. The l[subscript 1]-l[subscript 2] regularization and the cross validation generates an unbiased and objective output with a low classification error. We feel that the application of this algorithm to tumor biology will be instrumental to analyze gene expression signatures hidden in the transcriptome that, like hypoxia, may be major determinant of the course of the disease. |
| first_indexed | 2024-09-23T16:25:40Z |
| format | Article |
| id | mit-1721.1/52504 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:25:40Z |
| publishDate | 2010 |
| publisher | BioMed Central |
| record_format | dspace |
| spelling | mit-1721.1/525042022-09-29T19:47:38Z The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines The l1-l2 regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines Varesio, Luigi Verri, Alessandro Mosci, Sofia Barla, Annalisa Fardin, Paolo Rosasco, Lorenzo Andrea Massachusetts Institute of Technology. Center for Biological & Computational Learning McGovern Institute for Brain Research at MIT Rosasco, Lorenzo Andrea Rosasco, Lorenzo Andrea Background: Gene expression signatures are clusters of genes discriminating different statuses of the cells and their definition is critical for understanding the molecular bases of diseases. The identification of a gene signature is complicated by the high dimensional nature of the data and by the genetic heterogeneity of the responding cells. The l[subscript 1]-l[subscript 2] regularization is an embedded feature selection technique that fulfills all the desirable properties of a variable selection algorithm and has the potential to generate a specific signature even in biologically complex settings. We studied the application of this algorithm to detect the signature characterizing the transcriptional response of neuroblastoma tumor cell lines to hypoxia, a condition of low oxygen tension that occurs in the tumor microenvironment. Results: We determined the gene expression profile of 9 neuroblastoma cell lines cultured under normoxic and hypoxic conditions. We studied a heterogeneous set of neuroblastoma cell lines to mimic the in vivo situation and to test the robustness and validity of the l[subscript 1]-l[subscript 2] regularization with double optimization. Analysis by hierarchical, spectral, and k-means clustering or supervised approach based on t-test analysis divided the cell lines on the bases of genetic differences. However, the disturbance of this strong transcriptional response completely masked the detection of the more subtle response to hypoxia. Different results were obtained when we applied the l[subscript 1]-l[subscript 2] regularization framework. The algorithm distinguished the normoxic and hypoxic statuses defining signatures comprising 3 to 38 probesets, with a leave-one-out error of 17%. A consensus hypoxia signature was established setting the frequency score at 50% and the correlation parameter ε equal to 100. This signature is composed by 11 probesets representing 8 well characterized genes known to be modulated by hypoxia. Conclusion: We demonstrate that l[subscript 1]-l[subscript 2] regularization outperforms more conventional approaches allowing the identification and definition of a gene expression signature under complex experimental conditions. The l[subscript 1]-l[subscript 2] regularization and the cross validation generates an unbiased and objective output with a low classification error. We feel that the application of this algorithm to tumor biology will be instrumental to analyze gene expression signatures hidden in the transcriptome that, like hypoxia, may be major determinant of the course of the disease. EU Integrated Project Health-e-Child Fondazione Italiana per la Lotta al Neuroblastoma Foreign Investment Review Board (project RBIN04PARL) 2010-03-11T16:33:39Z 2010-03-11T16:33:39Z 2009-10 2009-05 Article http://purl.org/eprint/type/JournalArticle 1471-2164 http://hdl.handle.net/1721.1/52504 Fardin, Paolo et al. “The l1-l2 regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines.” BMC Genomics 10.1 (2009): 474. https://orcid.org/0000-0001-6376-4786 en_US http://dx.doi.org/10.1186/1471-2164-10-474 BMC Genomics Creative Commons Attribution http://creativecommons.org/licenses/by/2.0/ application/pdf BioMed Central BioMed Central |
| spellingShingle | Varesio, Luigi Verri, Alessandro Mosci, Sofia Barla, Annalisa Fardin, Paolo Rosasco, Lorenzo Andrea The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title | The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title_full | The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title_fullStr | The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title_full_unstemmed | The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title_short | The l[subscript 1]-l[subscript 2] regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| title_sort | l subscript 1 l subscript 2 regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines |
| url | http://hdl.handle.net/1721.1/52504 https://orcid.org/0000-0001-6376-4786 |
| work_keys_str_mv | AT varesioluigi thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT verrialessandro thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT moscisofia thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT barlaannalisa thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT fardinpaolo thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT rosascolorenzoandrea thelsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT varesioluigi thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT verrialessandro thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT moscisofia thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT barlaannalisa thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT fardinpaolo thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT rosascolorenzoandrea thel1l2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT varesioluigi lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT verrialessandro lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT moscisofia lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT barlaannalisa lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT fardinpaolo lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines AT rosascolorenzoandrea lsubscript1lsubscript2regularizationframeworkunmasksthehypoxiasignaturehiddeninthetranscriptomeofasetofheterogeneousneuroblastomacelllines |