DataSheet_1_Measurement of Conditional Relatedness Between Genes Using Fully Convolutional Neural Network.docx
Measuring conditional relatedness, the degree of relation between a pair of genes in a certain condition, is a basic but difficult task in bioinformatics, as traditional co-expression analysis methods rely on co-expression similarities, well known with high false positive rate. Complement with prior-knowledge similarities is a feasible way to tackle the problem. However, classical combination machine learning algorithms fail in detection and application of the complex mapping relations between similarities and conditional relatedness, so a powerful predictive model will have enormous benefit for measuring this kind of complex mapping relations. To this need, we propose a novel deep learning model of convolutional neural network with a fully connected first layer, named fully convolutional neural network (FCNN), to measure conditional relatedness between genes using both co-expression and prior-knowledge similarities. The results on validation and test datasets show FCNN model yields an average 3.0% and 2.7% higher accuracy values for identifying gene–gene interactions collected from the COXPRESdb, KEGG, and TRRUST databases, and a benchmark dataset of Xiao-Yong et al. research, by grid-search 10-fold cross validation, respectively. In order to estimate the FCNN model, we conduct a further verification on the GeneFriends and DIP datasets, and the FCNN model obtains an average of 1.8% and 7.6% higher accuracy, respectively. Then the FCNN model is applied to construct cancer gene networks, and also calls more practical results than other compared models and methods. A website of the FCNN model and relevant datasets can be accessed from https://bmbl.bmi.osumc.edu/FCNN.
History
References
- https://doi.org//10.1038/onc.2013.456
- https://doi.org//10.1371/journal.pone.0118731
- https://doi.org//10.2307/1270048
- https://doi.org//10.1038/s41598-019-42408-2
- https://doi.org//10.1093/biostatistics/kxt017
- https://doi.org//10.1038/nbt.3300
- https://doi.org//10.1093/nar/gks1193
- https://doi.org//10.1038/nbt.2601
- https://doi.org//10.1186/1471-2164-13-405
- https://doi.org//10.1186/gb-2006-7-2-r17
- https://doi.org//10.1093/nar/gkm323
- https://doi.org//10.1093/nar/gkh036
- https://doi.org//10.1093/nar/gkq1018
- https://doi.org//10.1093/nar/gkt1102
- https://doi.org//10.1093/nar/gki072
- https://doi.org//10.1093/bioinformatics/bti093
- https://doi.org//10.1038/ng.3421
- https://doi.org//10.1198/073500102753410444
- https://doi.org//10.1007/s10545-010-9128-0
- https://doi.org//10.1073/pnas.95.25.14863
- https://doi.org//10.1038/s41598-019-42408-2
- https://doi.org//10.1159/000493215
- https://doi.org//10.1001/jama.296.16.1958-d
- https://doi.org//10.1093/nar/gkx1013
- https://doi.org//10.1007/s11060-017-2680-9
- https://doi.org//10.1093/bioinformatics/bti268
- https://doi.org//10.1007/s13277-014-2766-3
- https://doi.org//10.1186/1471-2105-11-S11-S3
- https://doi.org//10.1002/0470857897.ch8
- https://doi.org//10.1093/nar/28.1.27
- https://doi.org//10.1186/1471-2105-11-S3-S9
- https://doi.org//10.1371/journal.pone.0101004
- https://doi.org//10.1371/journal.pone.0050411
- https://doi.org//10.1038/s41598-017-18705-z
- https://doi.org//10.1111/j.1466-8238.2007.00358.x
- https://doi.org//10.1038/ng.890
- https://doi.org//10.1016/j.molcel.2015.12.014
- https://doi.org//10.1016/j.ctrv.2016.03.005
- https://doi.org//10.1093/nar/gkr952
- https://doi.org//10.1186/1471-2105-9-327
- https://doi.org//10.1186/1755-8794-4-12
- https://doi.org//10.1016/0092-8674(91)90392-C
- https://doi.org//10.1038/sj.onc.1207392
- https://doi.org//10.1093/dnares/dsp016
- https://doi.org//10.1186/s12859-017-1561-8
- https://doi.org//10.1021/pr100618t
- https://doi.org//10.1007/s11063-017-9720-5
- https://doi.org//10.1002/gcc.20826
- https://doi.org//10.1155/2014/646193
- https://doi.org//10.1038/nature10350
- https://doi.org//10.1093/nar/gkv007
- https://doi.org//10.1093/nar/gku1042
- https://doi.org//10.1186/1471-2105-13-328
- https://doi.org//10.1214/aos/1074290335
- https://doi.org//10.1038/s41598-017-05728-9
- https://doi.org//10.1093/bioinformatics/btm087
- https://doi.org//10.1038/s41598-019-40780-7
- https://doi.org//10.1093/nar/28.1.289
- https://doi.org//10.1021/pr100618t
- https://doi.org//10.1038/s41598-019-40780-7
- https://doi.org//10.1038/nrc.2016.81
- https://doi.org//10.1038/ncomms4231
- https://doi.org//10.1093/nar/gku1163
- https://doi.org//10.1159/000439571
- https://doi.org//10.1002/prot.21018
- https://doi.org//10.1093/nar/gkw1119
- https://doi.org//10.1186/1471-2105-7-197
- https://doi.org//10.1209/0295-5075/87/38002
- https://doi.org//10.1109/ACCESS.2019.2908501
- https://doi.org//10.1109/ACCESS.2018.2890414
Usage metrics
Read the peer-reviewed publication
Categories
- Gene and Molecular Therapy
- Biomarkers
- Genetics
- Genetically Modified Animals
- Developmental Genetics (incl. Sex Determination)
- Epigenetics (incl. Genome Methylation and Epigenomics)
- Gene Expression (incl. Microarray and other genome-wide approaches)
- Livestock Cloning
- Genome Structure and Regulation
- Genetic Engineering
- Genomics