search

shenlab-sinai / ngsplot

Quick mining and visualization of NGS data by integrating genomic databases
Other
252 stars 65 forks source link

flat plots in ngsplots #69

Open ShivPet opened 7 years ago

ShivPet commented 7 years ago

HI Li, Need some urgent help. I have the same issue. My code was: ngs.plot.r -G hg19 -R tss -C P074A2_S5_R1_001.fastq.trim.align.BLMTfil.sort.rmdup.bam -O P074A1.tss -T P074A2TSS. The data is peaks called using MACS2. My plots are like so:

P074A1.tss.heatmap.pdf P074A1.tss.avgprof.pdf

Kindly help. Im a total novice and have no idea what to do.

Thanks Shiva.

amitpande74 commented 7 years ago

H shivPet, Can you tel us something more about your BAM file.....how did you process it ....sorting etc and if your data are the peaks from MACS2 then why are you using the -R tss flag ? You should be using ngs.plot.r -G hg19 -R bed -E your peak file.bed -C bam file -O P074A1.tss -T P074A2TSS. try it !

ShivPet commented 7 years ago

Hi Amitpande, Thanks so much for your prompt reply. My data is actually ATAC-seq data.I just wanted to initially see if my reads from my alignment are (slightly) enriched at the TSS sites- which is how it seems visually from IGV. The code for generating and processing my bam files are in the attached pdf ATAC-Seq pipeline-pdf.pdf. meanwhile ill try your suggestion using my peak files.

Thanks, Shiva

ADSL0067 commented 7 years ago

Hi, Amitpande I also get issue as same as ShivPet met. The bam file (GSM733659) was downloaded from ncbi which made by maq, sorted by samtools and remove the duplication by picard. ngs.plot can make the enrichment results using the default TSS which attached by the sofeware, but when use TSS.bed ( format: chr1\91104\91105\ENSG00000239945.1 ) which make by myselfe, it's no enrichment pattern. And the TSS.bed can working perfect with the bam made by bowtie. What should i do? MY code: ngs.plot2.r -G hg19 -R bed -E Gencode_CodingANDlincRNA.filter.bed -C GSM733659_hg19_Helas3Control.rmdup.bam -O GSM733659_hg19_Helas3Control -SC 0,1 -GO prod -LOW 0 -L 2000 -FL 100 -T GSM733659 -RR 60 -CO Coral

Any advice is welcome.

Thanks,

Liang

lishen commented 7 years ago

When you see a flat plot, the first thing to check is whether your bam files contain the same chromosome names as your references. If not, you can reheader the bam files.

Also, I just release 2.63 which fixes some critical issues. Be sure to use the new version.

sinirautio commented 7 years ago

Hi,

were you able to solve the problem? I have exactly the same problem, haven't had it previously but now I downloaded the data from NCBI and it just shows a flat plot even though I know there are reads aligning to those regions in the bed file.