Editing
Kun:LabNotes/MONOD/2014-8-12
(section)
Jump to navigation
Jump to search
Warning:
You are not logged in. Your IP address will be publicly visible if you make any edits. If you
log in
or
create an account
, your edits will be attributed to your username, along with other benefits.
Anti-spam check. Do
not
fill this in!
===4. Mix hap analysis=== *It looks like simple classification of a haplotype into blood or tumor based on the number of methylated CpG is a little bit crude, and it's hard to come up with a robust strategy for detecting tumor haplotypes in patient's plasma. *I decided to go back to the mix hap analysis and make a number of adjustments. **My previous version of mix hap analysis assumes BSPP data, in which most or all reads overlap perfectly. With RRBS data, many reads in a target region don't overlap. So I needed to be more flexible in comparing haplotypes. **The average read depth is lower in RRBS data. For CpG sites that have low or no coverage in normal plasma, I need to use the methylation level of whole blood WGBS data to estimate the haplotype frequency. **I decided to switch to LOD (log of odds ratio) between normal plasma and tumor. Essentially, I created two reference hapInfo file, one for all normal plasma (NC-P) samples combined, another for all primary tumor samples of one cancer type (PC, CC or LC) combined. For each sequence reads in the patient plasma, I asked what's the probability that the haplotype came from blood and the probability it came from tumor. Then I took the log of the odds ratio between these two probabilities, and sum them up for each target region and for the entire genome. *I started with all RRBS targets within the UMRs, but then need to trim the target list, based on which target is informative. Since the sample size is limited, doing cross validation is not realistic. So I simply pooled all patient plasma data (treating PC, CC, LC separately), and ran mixMethHapAnalysis_19Aug14.pl against pooled normal plasma and pooled tumor to identify the regions that have the highest LOD scores. ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt 6-T-ALL.hapInfo.txt 6-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_CC.txt & ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt 7-T-ALL.hapInfo.txt 7-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_LC.txt & ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt PC-T-ALL.hapInfo.txt PC-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_PC.txt & *To evaluate the level of false positives due to variability in normal plasma, I took one normal plasma as a test sample to compare with the pool of the other seven normal plasma. I did this for each of the 8 plasma samples, and reported a list of regions that show high LOD (false positives). These regions were excluded from the list of informative markers. It appeared that whether there are false positive calls in a region also depends on which tumor reference file is used. For now, I just identified these regions for each of the three tumor types separately. In the future, we might need to process a lot of normal plasma and primary tumor samples to fully cover the biological variability. ./find_NC-P_high-LOD_targets.pl > NC-P_high-LOD_targets.txt *I manually split the NC-P_high-LOD_targets.txt file into three based on the tumor types, then extracted the informative marker lists based on the differences of LOD scores between patient plasma and normal plasma. ./filter_targets.pl RRBS_targets_LOD_PC.txt NC-P_high-PC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_PC_top_347.txt ./filter_targets.pl RRBS_targets_LOD_LC.txt NC-P_high-LC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_LC_top_346.txt ./filter_targets.pl RRBS_targets_LOD_CC.txt NC-P_high-CC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_CC_top_302.txt *Then I searched for the tumor haplotypes in each plasma samples within the marker regions. ./get_all_CC-P_LOD.pl > CC-P_LOD_out.txt & ./get_all_LC-P_LOD.pl > LC-P_LOD_out.txt & ./get_all_PC-P_LOD.pl > PC-P_LOD_out.txt & [[Image:Colon_cancer_UMR_tumor_haplotype_load.png|350px]][[Image:Lung_cancer_UMR_tumor_haplotype_load.png|350px]][[Image:Pancreatic_cancer_UMR_tumor_haplotype_load.png|350px]] *Some obervations: **Tumor loads are detectible in patient's plasma, and the load varies from one sample to another, more so than from one tumor type to another. **I was only able to achieve this after removing all regions that contain false-positive haplotypes in normal plasma. Cross validation is not possible with so few normal plasma samples. So we need more normal plasma data to have a real assessment of the false positive rate. **Are there the regions that contain tumor haplotypes across multiple samples?? A total 241 regions were found in 2 or more samples. These regions are enriched for DNA binding proteins, especially HOX genes. [[Image:Common_high_LOD_regions_InterPro_enrichment.png|800px]] [[Image:Common_high_LOD_regions_GO_MolecularFunction_enrichment.png|800px]] **The regions that appeared in 5 or more samples are listed below. {| {{table}} | align="center" style="background:#f0f0f0;"|'''Target_region''' | align="center" style="background:#f0f0f0;"|'''# samples''' | align="center" style="background:#f0f0f0;"|'''Gene''' |- | chr16:29675696-29676014||13||SPN |- | chr11:72295386-72295938||12||PDE2A |- | chr16:67428179-67429159||12||ZDHHC1 |- | chr17:76127129-76128704||12||TMC6 |- | chr2:177016416-177016749||12||HOXD4 |- | chr19:1074331-1075173||10||HMHA1 |- | chr6:100897213-100897619||10||SIM1 |- | chr2:127839497-127839766||8||BIN1 |- | chr22:20784703-20786333||8||SCARF2 |- | chr6:42071960-42072652||8||C6orf132 |- | chr12:124941780-124941976||7||NCOR2 |- | chr19:18980616-18981821||7||GDF1/CERS1 |- | chr2:177036288-177036973||7||HOXD3 |- | chr22:19710616-19712149||7||GP1BB |- | chr11:555351-556680||6||LMNTD2 |- | chr12:49942621-49943178||6||KCNH3 |- | chr14:37127773-37128202||6||PAX9 |- | chr16:54970220-54970503||6||CRNDE/IRX5 |- | chr17:5404000-5404600||6||LOC728392 |- | chr2:111876191-111877448||6||ACOXL |- | chr4:169798910-169799623||6||PALLD |- | chr7:127671170-127671464||6||SND1 |- | chr7:155246283-155248093||6||EN2 |- | chr8:80804055-80804317||6||GDF6 |- | chr8:97156647-97158028||6||GDF6 |- | chr10:118030474-118031344||5||GFRA1 |- | chr1:151810190-151811682||5||C2CD4D |- | chr12:115105189-115105423||5||TBX3 (downstream) |- | chr1:225865165-225865313||5||ENAH |- | chr13:112547537-112548375||5||LINC00354 |- | chr15:96895306-96895504||5||NR2F2 (downstream) |- | chr16:86612197-86613526||5||FOXL1 |- | chr17:70111952-70112499||5||SOX9-AS1 |- | chr1:91300884-91301262||5||#not near a gene# |- | chr19:4328686-4329420||5||STAP2 |- | chr21:36900927-36901032||5||LOC100506403 |- | chr2:171678137-171678305||5||GAD1 |- | chr2:177023686-177024305||5||HOXD3/4 |- | chr2:233925159-233925345||5||INPP5D |- | chr5:88184868-88185224||5||MEF2C |- | chr6:10385059-10385426||5||TFAP2A |- | chr8:142528135-142529145||5||MROH5 |- | |} *A region related to SPN gene Normal Plasma Primary Tumor(LC-4) Matched Patient Plasma(LC-4) [[Image:chr16.29675696-29676014_SPN_NCP.png|400px]][[Image:chr16.29675696-29676014_SPN_LC-T4.png|400px]][[Image:chr16.29675696-29676014_SPN_LC-P4.png|200px]] ====Repeat the analysis using a refined set of targets==== *The analyses above were performed based on a set of RRBS targets compiled based on the read depth of CpG sites in tumor data. Some of the target regions are fairly large, and contain many non-overlapping reads that cannot be directly compared in the haplotype analysis. *A different way to define the RRBS targets is based on the read depth distribution (not just CpG sites), which might better separate non-overlapping reads. *To test this idea, I compiled a new set of RRBS targets with the following steps. samtools merge /home/kunzhang/CpgMIP/Data/MONOD/1407-combined_RRBS/Primary_tumor_ALL.merged.bam 6-T-1_1.merged.bam 6-T-1_2.merged.bam 6-T-2_1.merged.bam 6-T-2_2.merged.bam 6-T-3_1.merged.bam 6-T-3_2.merged.bam 6-T-4_1.merged.bam 6-T-4_2.merged.bam 6-T-5_1.merged.bam 6-T-5_2.merged.bam 7-T-1_1.merged.bam 7-T-1_2.merged.bam 7-T-2_1.merged.bam 7-T-2_2.merged.bam 7-T-3_1.merged.bam 7-T-3_2.merged.bam 7-T-4_1.merged.bam 7-T-4_2.merged.bam 7-T-5_1.merged.bam 7-T-5_2.merged.bam PC-T-1_1.merged.bam PC-T-1_2.merged.bam PC-T-2_1.merged.bam PC-T-2_2.merged.bam PC-T-4_1.merged.bam PC-T-4_2.merged.bam PC-T-6_1.merged.bam PC-T-6_2.merged.bam PC-T-7_1.merged.bam PC-T-7_2.merged.bam samtools sort Primary_tumor_ALL.merged.bam Primary_tumor_ALL.sorted & nohup /home/kunzhang/softwares/bedtools-2.17.0/bin/bedtools genomecov -bg -split -ibam Primary_tumor_ALL.sorted.bam > Pimary_tumor_ALL.genomecov.bed & awk '$4>50 { print $1"\t"$2"\t"$3}' Primary_tumor_ALL.genomecov.bed | /home/kunzhang/softwares/bedtools-2.17.0/bin/bedtools merge -d 10 -i - > Primary_tumor_ALL.genomecov.RD50.merged.bed awk '$3-$2>80 {print $1"\t"$2"\t"$3"\t"$3-$2+1}' Primary_tumor_ALL.genomecov.RD50.merged.bed> Primary_tumor_ALL.genomecov.RD50_80UP.merged.bed sh [[media:1407-combined_expanded_31Aug14_step1_batch_command.txt|1407-combined_expanded_31Aug14_step1_batch_command.sh]] & ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt 6-T-ALL.hapInfo.txt 6-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_CC.txt & ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt 7-T-ALL.hapInfo.txt 7-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_LC.txt & ../../mixMethHapAnalysis_19Aug14.pl NC-P-ALL.hapInfo.txt PC-T-ALL.hapInfo.txt PC-P-ALL.hapInfo.txt /home/kunzhang/CpgMIP/MONOD/Public_data/WB_middle_age_UMR.BED.txt | sort -k 2,2nr > RRBS_targets_LOD_PC.txt & ./find_NC-P_high-LOD_targets.pl > NC-P_high-LOD_targets.txt ./filter_targets.pl RRBS_targets_LOD_PC.txt NC-P_high-PC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_PC_top_1914.txt ./filter_targets.pl RRBS_targets_LOD_LC.txt NC-P_high-LC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_LC_top_1629.txt ./filter_targets.pl RRBS_targets_LOD_LC.txt NC-P_high-CC-LOD_targets.txt | sort -k 2,2nr > RRBS_targets_LOD_CC_top_1487.txt ./get_all_CC-P_LOD.pl > CC-P_LOD_out.txt & ./get_all_LC-P_LOD.pl > LC-P_LOD_out.txt & ./get_all_PC-P_LOD.pl > PC-P_LOD_out.txt & *A total of 822 regions that contain cancer haplotypes in two or more samples were identified. [[Image:Colon_cancer_UMR_tumor_haplotype_load_31Aug14.png|350px]][[Image:Lung_cancer_UMR_tumor_haplotype_load_31Aug14.png|350px]][[Image:Pancreatic_cancer_UMR_tumor_haplotype_load_31Aug14.png|350px]] [[Image:CC_LC_PC_common_Go_biological_process_31Aug14.png|1200px]] [[Image:CC_LC_PC_common_Go_molecular_function_31Aug14.png|1200px]] [[Image:CC_LC_PC_common_InterPro_31Aug14.png|800px]] [[Image:CC_LC_PC_common_MSigDB_cancer_neighborhood_31Aug14.png|800px]]
Summary:
Please note that all contributions to ZhangLabWiki may be edited, altered, or removed by other contributors. If you do not want your writing to be edited mercilessly, then do not submit it here.
You are also promising us that you wrote this yourself, or copied it from a public domain or similar free resource (see
ZhangLabWiki:Copyrights
for details).
Do not submit copyrighted work without permission!
Cancel
Editing help
(opens in new window)
Navigation menu
Personal tools
Not logged in
Talk
Contributions
Create account
Log in
Namespaces
Page
Discussion
English
Views
Read
Edit
View history
More
Search
Navigation
Main Page
Current events
Recent changes
Random page
Investigators
Matt Cai
Song Chen
Eric Chu
Dinh Diep
Elizabeth Duong
Shicheng Guo
Alan Fung
Daniel Jacobsen
Blue Lake
Huy Lam
Alice Li
Andrew Richards
Brandon Sos
Chris Wei
Yan Wu
Kun Zhang
Tools
What links here
Related changes
Special pages
Page information