We developed a novel algorithm CARMEN to identify noncoding variants which function as the disease-risk allele through modulating gene expression. Here, we provide online predict services to all users.
You can upload a VCF file or paste the content to run a new CARMEN analysis. Here is the description of of the VCF file or the content you paste:
|Chr||Chromosome, necessary. Please use the format like chr16 or chrX.|
|Rs||Reference SNP ID, unnecessary. If none, please use . as a placeholder.|
|Ref||Ref allele, necessary. It should be AGCT or its combination. Indel should be shorter than 50.|
|Alt||Alt allele, necessary. It should be AGCT or its combination. Indel should be shorter than 50.|
|Comment||More information about the variant, unnecessary. If none, please use . as a placeholder.|
You can also download example here. Now, the maximum number of your data is 1000.
When you finish uploading or pasting, you should choose the reference genome version of your data (GRCh38 or GRCh37). Besides, you can also enter your e-mail for receiving the update of the task state. If you submit task successfully, you will get a task id and the page will jump to task state search page in 3 seconds automatically.
Use the task id, you can check the state of the task you submitted through CARMEN. The page will refresh automatically. For a submitted task, there are five kinds of states:
|Running||The task is running, please wait patiently.|
|Finished||The task finished successfully.|
|Finished with warning||The task finished successfully. However, part of your variants cannot be predicted. You can check the task report for more details.|
|Unsuccessful||The task cannot be finished. Please try other variants. You can check the task report for more details.|
|ERROR||Please feel free to contact us with the task id, for the reason that we can deal with it as soon as possible.|
When the task finishs, the results will be stored for 7 days.
When the task finishs successfully, you can use task id to fetch the result. Following information will be displayed:
|Pos||Position. Reference genome version is GRCh38.|
|Rs||Reference SNP ID.|
|CARMEN-E||This score indicates the variant effect on gene expression level.|
|CARMEN-F||This score indicates the risk probability of the variant.|
|CARMEN-Score||This score combines the above two scores.|
|Comment||More information about the variant.|
|Details||You can click it for more details about the result.|
If you click "details" for more details about the task, the following information will be displayed:
Overall information: the summary information of the result, and you can check the position of the variant in UCSC genome browser through related link.
CARMEN-Score: it shows target CARMEN-Score and the score distribution of the test set used in model training.
Annotation: it contains two parts, which includes chromatin profile features, DNA physicochemical properties and evolutionary features.
Besies, you can also download the predict result(.csv) and annotations file(.txt).
You can download the standalone version of CARMEN through here.
Team: Fangyuan Shi, Yu Wang, Dong Huang, Yu Liang, Nan Liang, Xiaowei Chen, Ge Gao.
Cite: Shi, F.Y., Wang, Y., Huang, D., Liang, Y., Liang, N., Chen, X.W., and Gao, G. (2021). Computational Assessment of the Regulation-Modulating Potential for Noncoding Variants. Genomics, Proteomics & Bioinformatics.