Matt:LabNotes/2016-4-8

From ZhangLabWiki
Jump to navigation Jump to search

Design Padlock Probe Arms for Mouse and Human Housekeeping Gene Probe Set[edit]

  • For DARTFISH in human and mouse tissues, we are most interested in seeing the heterogeneity and variation in gene expression
  • However, to distinguish biological variation from technical noise it would be nice to have some probes for genes that are uniformly detected in each tissue
  • Files: Dropbox\GradZhangLab\CA12k_Apr2016\ControlGenes\MouseEmbryo and Dropbox\GradZhangLab\CA12k_Apr2016\ControlGenes\HumanBrain

Mouse Embryo[edit]

Gene Selection[edit]

  • "high_genes_mouse_project.xls" contains 54 high expression genes with lowest variance (variance included)
  • Selected top 25 lowest variance genes "Top25_LowVarGenes.txt"
    • Includes Gapdh and Actg
    • After going through process resulted in <200 probes so expanded genes
  • Selected all genes not mitochondrial (mt-)
    • 42 genes left: "42genes.txt"
      • Changed Gm5506 -> Eno1b (MGI symbol)

Get Transcript Sequences[edit]

  • Biomart browser interface
 Dataset
 Mus musculus genes (GRCm38.p4)
 Filters
 with MGI ID(s): Only
 MGI symbol [e.g. Mir1901]: [ID-list specified]
 Status (gene): KNOWN
 Status (transcript): KNOWN
 Attributes
 Ensembl Gene ID
 Ensembl Transcript ID
 Chromosome Name
 Exon Rank in Transcript
 Exon Chr Start (bp)
 Exon Chr End (bp)
 Strand
 Associated Gene Name
  • 42 Unique Ensembl Gene IDs and Associated Gene Names

Create ppDesigner Target Files[edit]

  • Files in genome-miner:~/scratch/CA12kApr2016_ProbeDesign/ControlGenes/MouseEmbryo
  • Use CreateTargetFile_contig.pl to create target file where targets are contigs of exons
    • Script is modified from here
  • Sort target files into each chromosome and remove 25bp from each end of target and switch strand
    • SortTargetFilesByChr.pl

Run ppDesigner[edit]

ppDesignerCommands.sh

 #!/bin/bash
 for indx in 1 2 3 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 X
 do
 /home/mzcai/scratch/CA12kApr2016_ProbeDesign/ControlGenes/MouseEmbryo/opt/ppDesigner/src/ppDesigner.pl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/ControlGenes/MouseEmbryo/jobFile_chr$indx.pl > /home/mzcai/scratch/CA12kApr2016_ProbeDesign/ControlGenes/MouseEmbryo/outputFile_chr$indx.txt &
 wait
 done
  • Add target base to arm with lowest Tm to create zero-gap padlock probe
  • Also filter out any probes targeting soft-masked regions (indicated by lowercase reference sequence)
    • ConvertToZeroGapProbe.pl
 perl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/ConvertToZeroGapProbe.pl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/outputFile_chr$indx.txt > /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/outputFile_0gap_chr$indx.txt &
 cat outputFile_0gap_chr*.txt > outputFile_0gap.txt
  • 258 probes
  • 191 exons
  • 41 genes
    • Missing Ubc

perl Probes2fasta.pl < outputFile_0gap.txt > outputFile_0gap.fa

 /home/kunzhang/softwares/Novocraft/novocraft/novoalign -d /home/mzcai/scratch/Genomes/mm10/novoalign/mm10_refMrna.ndx -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_mm10refMrna.out &
 /home/kunzhang/softwares/Novocraft/novocraft/novoalign -d /home/mzcai/scratch/Genomes/mm10/novoalign/mm10.ndx -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_mm10.out &

perl CleanupProbelist.pl Remove probes that did not align to refMrna or had multiple alignments to mm10

  • 85 probes
  • 76 exons
  • 33 genes
  • Output: outputFile_0gap_filtered.txt

Human Brain[edit]

Gene Selection[edit]

  • Got TPM table of Blue's single neurons including unclassified ones (4,053 cells)
    • Google Drive: BA8 Analyses/TPM Matrices/"09.anno_exon_log2tpm.dat_nc"
  • Calculated variance and mean for each gene (Dropbox/PythonScripts/CalcTPMmatrix.py) -> "neurons_variance.txt"
  • Filter only genes with mean TPM >= 5
  • Sort by variance low to high
      1. 1 gene is MALAT1
    • Other top genes are KIF5C and SNAP25
    • I recognize these in my current CA12kNov2014_V4 probe set
  • Take Top 200 genes with lowest variance "200genes.txt"
    • Change LPHN1 -> ADGRL1 (updated HGNC symbol)

Get Transcript Sequences[edit]

  • Biomart browser interface
 Dataset
 Homo sapiens genes (GRCh38.p5)
 Filters
 with HGNC ID(s): Only
 HGNC symbol(s) [e.g. NTN3]: [ID-list specified]
 Status (gene): KNOWN
 Status (transcript): KNOWN
 Attributes
 Ensembl Gene ID
 Ensembl Transcript ID
 Chromosome Name
 Exon Rank in Transcript
 Exon Chr Start (bp)
 Exon Chr End (bp)
 Strand
 Associated Gene Name
  • Remove coordinates on PATCH (fixes and novel), HSCHR (haplotype variants), and chromosomes
    • 198 Unique Ensembl Gene IDs and 198 Unique HGNC Symbols
      • Could not find LZTS3 and RP11-217O12.1

Create ppDesigner Target Files[edit]

  • Files in genome-miner:~/scratch/CA12kApr2016_ProbeDesign/HumanBrain
  • Use CreateTargetFile_contig.pl to create target file where targets are contigs of exons
    • Script is modified from here
  • Sort target files into each chromosome and remove 25bp from each end of target and switch strand
    • SortTargetFilesByChr.pl

Run ppDesigner[edit]

ppDesignerCommands.sh

 #!/bin/bash
 for indx in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 X
 do
 /home/mzcai/scratch/CA12kApr2016_ProbeDesign/MouseEmbryo/opt/ppDesigner/src/ppDesigner.pl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/MouseEmbryo/jobFile_chr$indx.pl > /home/mzcai/scratch/CA12kApr2016_ProbeDesign/MouseEmbryo/outputFile_chr$indx.txt &
 wait
 done
  • Add target base to arm with lowest Tm to create zero-gap padlock probe
  • Also filter out any probes targeting soft-masked regions (indicated by lowercase reference sequence)
    • ConvertToZeroGapProbe.pl
 perl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/ConvertToZeroGapProbe.pl /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/outputFile_chr$indx.txt > /home/mzcai/scratch/CA12kApr2016_ProbeDesign/HumanEmbryo/outputFile_0gap_chr$indx.txt &
 cat outputFile_0gap_chr*.txt > outputFile_0gap.txt
  • 7,743 probes
  • 4,756 exons
  • 196 genes
    • Removed MT- gene and missing RN7SL2

perl Probes2fasta.pl < outputFile_0gap.txt > outputFile_0gap.fa

 home/kunzhang/softwares/Novocraft/novocraft/novoalign -d /home/kunzhang/RNAseq/Data/CommonFiles/refMrna.ndx -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_refMrna.out &
 /home/kunzhang/softwares/Novocraft/novocraft/novoalign -d /home/kunzhang/softwares/Novocraft/novocraft/human_g1k_v37 -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_human_g1k_v37.out &

perl CleanupProbelist.pl Remove probes that did not align to refMrna or had multiple alignments to hg38

  • 5,569 probes
  • 3,772 exons
  • 191 genes
    • Missing PWAR6, SNORD116-8, SNORD116-15, SNORD116-16, SNORD116-24
  • Output: outputFile_0gap_filtered.txt