Abstract
Supervised machine learning is a powerful and widely used method for analyzing high-content screening data. Despite its accuracy, efficiency, and versatility, supervised machine learning has drawbacks, most notably its dependence on a priori knowledge of expected phenotypes and time-consuming classifier training. We provide a solution to these limitations with CellCognition Explorer, a generic novelty detection and deep learning framework. Application to several large-scale screening data sets on nuclear and mitotic cell morphologies demonstrates that CellCognition Explorer enables discovery of rare phenotypes without user training, which has broad implications for improved assay development in high-content screening.
References
46
Referenced
85
{'key': 'B1', 'first-page': '36', 'volume': '11', 'author': 'Abramoff MD', 'year': '2004', 'journal-title': 'Biophoton Int'}
/ Biophoton Int by Abramoff MD (2004)10.1126/science.1140324
/ Science by Bakal C (2007)10.1093/bioinformatics/17.12.1213
/ Bioinformatics by Boland MV (2001)10.1016/j.cell.2015.11.007
/ Cell by Boutros M (2015)10.1186/gb-2006-7-10-r100
/ Genome Biol by Carpenter AE (2006)10.1083/jcb.200910105
/ J Cell Biol by Conrad C (2010)10.1038/nature18610
/ Nature by Cuylen S (2016)10.1083/jcb.200101089
/ J Cell Biol by Daigle N (2001){'key': 'B11', 'first-page': '2121', 'volume': '12', 'author': 'Duchi J', 'year': '2011', 'journal-title': 'J Mach Learn Res'}
/ J Mach Learn Res by Duchi J (2011){'key': 'B12', 'first-page': '1', 'author': 'Durr O', 'year': '2016', 'journal-title': 'J Biomol Screen'}
/ J Biomol Screen by Durr O (2016)10.1186/1471-2105-14-292
/ BMC Bioinformat by Failmezger H (2013)10.1186/gb-2004-5-10-r80
/ Genome Biol by Gentleman RC (2004)10.1126/science.1141314
/ Science by Goshima G (2007)10.1016/j.cell.2012.06.039
/ Cell by Gudjonsson T (2012)10.1007/978-94-015-3994-4
/ Identification of Outliers. Monographs on Statistics and Applied Probability by Hawkins DM (1980)10.1038/nmeth.1486
/ Nat Methods by Held M (2010)10.1126/science.1127647
/ Science by Hinton GE (2006)10.1073/pnas.0808843106
/ Proc Natl Acad Sci USA by Jones TR (2009)10.1093/bioinformatics/btw252
/ Bioinformatics by Kraus OZ (2016)10.15252/msb.20177551
/ Mol Syst Biol by Kraus OZ (2017)10.1038/nature14539
/ Nature by LeCun Y (2015)10.1016/j.cell.2014.04.029
/ Cell by Liberali P (2014){'key': 'B26', 'first-page': '49', 'volume': '2', 'author': 'Mahalanobis PC', 'year': '1936', 'journal-title': 'Proc Natl Inst Sci India'}
/ Proc Natl Inst Sci India by Mahalanobis PC (1936)10.5334/jors.bg
/ J Open Res Software by Manning S (2014)10.1016/j.tcb.2016.03.008
/ Trends Cell Biol by Mattiazzi Usaj M (2016)10.1186/1471-2105-11-30
/ BMC Bioinformat by Misselwitz B (2010)10.1023/B:VLSI.0000003028.71666.44
/ J VLSI Sig Proc Syst by Murphy RF (2003){'key': 'B31', 'first-page': '372', 'volume': '27', 'author': 'Nesterov Y', 'year': '1983', 'journal-title': 'Soviet Mathematics Doklady'}
/ Soviet Mathematics Doklady by Nesterov Y (1983)10.1038/nmeth876
/ Nat Methods by Neumann B (2006)10.1038/nature08869
/ Nature by Neumann B (2010)10.1016/j.sigpro.2013.12.026
/ Signal Process by Pimentel MAF (2014)10.1038/nmeth.2097
/ Nat Methods by Rajaram S (2012)10.1093/bioinformatics/btp524
/ Bioinformatics by Ramo P (2009){'key': 'B38', 'first-page': '8614', 'author': 'Sainath TN', 'year': '2013', 'journal-title': 'In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)'}
/ In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) by Sainath TN (2013)10.1038/nmeth.2019
/ Nat Methods by Schindelin J (2012)10.1007/978-1-60327-993-2_7
/ Methods Mol Biol by Schmitz MH (2009)10.1038/ncb2092
/ Nat Cell Biol by Schmitz MH (2010)10.1162/089976601750264965
/ Neural Comput by Scholkopf B (2001)10.1242/jcs.123604
/ J Cell Sci by Sommer C (2013)10.1093/bioinformatics/btt175
/ Bioinformatics by Sommer C (2013){'key': 'B45', 'first-page': '1701', 'author': 'Taigman Y', 'year': '2014', 'journal-title': 'Proc CVPR IEEE'}
/ Proc CVPR IEEE by Taigman Y (2014){'key': 'B46', 'first-page': '774', 'volume': '24', 'author': 'Vapnik V', 'year': '1963', 'journal-title': 'Autom Remote Control'}
/ Autom Remote Control by Vapnik V (1963){'key': 'B47', 'first-page': '3371', 'volume': '11', 'author': 'Vincent P', 'year': '2010', 'journal-title': 'J Mach Learn Res'}
/ J Mach Learn Res by Vincent P (2010)10.1038/ncb2764
/ Nat Cell Biol by Yin Z (2013)10.1186/1471-2105-9-264
/ BMC Bioinformatics by Yin Z (2008)10.1038/nmeth.2046
/ Nat Methods by Zhong Q (2012)
Dates
Type | When |
---|---|
Created | 7 years, 10 months ago (Sept. 27, 2017, 8:50 p.m.) |
Deposited | 6 years, 1 month ago (July 14, 2019, 12:32 p.m.) |
Indexed | 1 month, 1 week ago (July 11, 2025, 9:18 p.m.) |
Issued | 7 years, 9 months ago (Nov. 7, 2017) |
Published | 7 years, 9 months ago (Nov. 7, 2017) |
Published Print | 7 years, 9 months ago (Nov. 7, 2017) |
@article{Sommer_2017, title={A deep learning and novelty detection framework for rapid phenotyping in high-content screening}, volume={28}, ISSN={1939-4586}, url={http://dx.doi.org/10.1091/mbc.e17-05-0333}, DOI={10.1091/mbc.e17-05-0333}, number={23}, journal={Molecular Biology of the Cell}, publisher={American Society for Cell Biology (ASCB)}, author={Sommer, Christoph and Hoefler, Rudolf and Samwer, Matthias and Gerlich, Daniel W.}, editor={Boone, Charles}, year={2017}, month=nov, pages={3428–3436} }