search

mquinodo / OFF-PEAK

CNV detection tool for WES data
6 stars 0 forks source link

ref genome #2

Closed FAFUshiyan closed 5 months ago

FAFUshiyan commented 5 months ago

The error occurs because I am using the hs37d5.fa reference genome, which does not have "chr" in the chromosome names.

Feature (chrX:154976443-154996443) beyond the length of chrX size (0 bp). Skipping. Feature (chrX:155026269-155046269) beyond the length of chrX size (0 bp). Skipping. Feature (chrX:155076095-155096095) beyond the length of chrX size (0 bp). Skipping.

mquinodo commented 5 months ago

Hello, Thank you for reporting the issue. I will adapt the scripts so they work with such reference without "chr". I guess in that case your BED file with the targeted regions also has no "chr"? Best, Mathieu

FAFUshiyan commented 5 months ago

Hello Mathieu, Thank you for your response. Yes, both my BED file and BAM files do not include "chr".

I have some additional questions I would like to ask:

  1. In your article, you mentioned that a minimum of 15 control samples is required. Should these 15 control samples be included in the list_BAM.txt file along with the case samples?
  2. When using the OFF-PEAK software, what specific processing steps are required for all BAM files?

Thank you for your help.

Best regards,

发件人: Mathieu Quinodoz @.> 日期: 星期一, 2024年4月15日 15:04 收件人: mquinodo/OFF-PEAK @.> 抄送: ShiYan @.>, Author @.> 主题: Re: [mquinodo/OFF-PEAK] ref genome (Issue #2)

Hello, Thank you for reporting the issue. I will adapt the scripts so they work with such reference without "chr". I guess in that case your BED file with the targeted regions also has no "chr"? Best, Mathieu

― Reply to this email directly, view it on GitHubhttps://github.com/mquinodo/OFF-PEAK/issues/2#issuecomment-2055733254, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASJOBNB4MYLLRWRQB76KY63Y5N3WXAVCNFSM6AAAAABGFJLAG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJVG4ZTGMRVGQ. You are receiving this because you authored the thread.Message ID: @.***>

mquinodo commented 5 months ago

Hello,

I have added a --nochr option for the first script (01_targets-processing.sh) and adapted 02_coverage-count.sh. The output of the second script will add the "chr" and step 3 (03_OFF-PEAK.R) can be performed normally.

  1. OFF-PEAK requires at least 15 samples to be run in a batch. When analyzing sample 1, it will take the 14 others as "controls". So you can simply put all the samples in the list_BAM.txt file (cases and controls or only cases).

  2. I think I do not understand your question. Could you explain more?

Best, Mathieu

FAFUshiyan commented 5 months ago

Hello Mathieu,

Thank you for the updates on the scripts and clarifying the process regarding the inclusion of control samples in the list_BAM.txt file.

Regarding the processing of BAM files, my question pertains to whether the following three steps are necessary:

  1. samtools fixmate
  2. GATK MarkDuplicates
  3. GATK ApplyBQSR

I appreciate your assistance in clarifying this matter.

Best regards, Yan

发件人: Mathieu Quinodoz @.> 日期: 星期二, 2024年4月16日 16:37 收件人: mquinodo/OFF-PEAK @.> 抄送: ShiYan @.>, Author @.> 主题: Re: [mquinodo/OFF-PEAK] ref genome (Issue #2)

Hello,

I have added a --nochr option for the first script (01_targets-processing.sh) and adapted 02_coverage-count.sh. The output of the second script will add the "chr" and step 3 (03_OFF-PEAK.R) can be performed normally.

  1. OFF-PEAK requires at least 15 samples to be run in a batch. When analyzing sample 1, it will take the 14 others as "controls". So you can simply put all the samples in the list_BAM.txt file (cases and controls or only cases).
  2. I think I do not understand your question. Could you explain more?

Best, Mathieu

― Reply to this email directly, view it on GitHubhttps://github.com/mquinodo/OFF-PEAK/issues/2#issuecomment-2058547679, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASJOBNEDPUGEXK6L2QDVR5TY5TPM7AVCNFSM6AAAAABGFJLAG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJYGU2DONRXHE. You are receiving this because you authored the thread.Message ID: @.***>

mquinodo commented 5 months ago

Hello,

These steps are not important as they will likely not influence the number of reads.

Best, Mathieu