Help and documentation
1. Use an example
Step 1: Choose an example from the pull-down menu. You can click "download example" to see how the example you chose looks like.
Step 2: Choose a model.
Step 3: Click "Submit".
2. Use your own data
1. The required format is as follows:
The input should be tab-delimited. The first coloum is chromosome. The second column is the starting position of a CpG site and the third column is the ending position. The forth and the fifth column is methylation count and sequencing depth, respectively. If methylation count or depth is not available, delete the row or use "NA" instead.
2. If you don't know how to get an input file:
If you have the output files of bismark produced by bismark_methylation_extractor, download bismark2bed.py and use this command:
python bismark2bed.py XXX.bismark.cov.gz outputFile.bed
XXX.bismark.cov.gz is one of the output files of bismark_methylation_extractor, and outputFile.bed is the output file of bismark2bed.py. You can use this file as the input for ctMethTracer.
If you have the output files of BSMAP produced by methratio.py, download bsmap2bed.py and use this command:
python bsmap2bed.py XXX_methratio.txt outputFile.bed
XXX_methratio.txt is one of the output files of methratio.py, and outputFile.bed is the output file of bsmap2bed.py. You can use this file as the input for ctMethTracer.
3. If you know how to get an input file:
Step 1: Click "Select file" to upload your own data. You can click "Example data" to see input format.
Step 2: Choose a model.
Step 3: (Optional) If you want to be notified by Email when your job is completed, input your Email address.
Step 4: Click "Submit".
3. Query your job
1. JobID: You can use this id to query your result.
2. Model: The model you chose.
3. Predicted type
Your data are predicted as this type according to the methylation level of cell-free DNA of your data. If the predicted type is normal, the body map would not be shown.
Multiple tumor types are included in our model. The scores assigned to each tumor type calculated by the model you chose are shown in the table. All scores are non-negative. The bigger the score is, the more the sample you provided is likely to have that kind of tumor. The tumor type with the biggest score is the predicted type.
The body map on the right is correlated with the table. Each tumor type corresponds to an organ in the body map. As you move your mouse onto a tumor type, the corresponding organ in the body map is highlighted, and vice versa.
At the bottom you can see the proportion of tumor-derived cell-free DNA in your data.
4. Score distribution
This graph shows the distribution of scores in TCGA cohort. X-axis represents intervals of scores, and Y-axis represents frequency of scores. Each point shows the frequency of scores in a specific interval.
5. Location of CpG sites
This chart illustrates the location of CpG sites in your data.
5. Where is the job ID
1. Your job id will be here right after you submit your job.
2. If you input you Email address before submit your job, you will receive an Email with your job id after submission.