Matt:LabNotes/2016-10-14

From ZhangLabWiki
Jump to navigation Jump to search

New Padlock Probes For RNA Capture with SplintR: CA12k_Oct2016[edit]

Human Brain[edit]

 Probe2padlockFISSEQ_Oct2016_HumanBrain_6Cycle_V4.pl > padlockFile_0gap_HumanBrain_V4.txt

Mouse Embryo[edit]

 perl Probe2padlockFISSEQ_Oct2016_MouseEmbryo_6Cycle_V6.pl > padlockFile_0gap_MouseEmbryo_V6.txt

Mouse Brain[edit]

Gene Selection[edit]

    • 174 new genes provided by Zizhen
    • 174 new genes: /GradZhangLab/DARTFISH Collaborations/Mouse Brain/Genelist_174genes_part2.txt
      • Gpr133 -> Adgrd1 (MGI Symbol)
      • A730090H04 -> Dlx4os (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
  • 173 Unique Ensembl Gene IDs and Associated Gene Names
    • Missing 5033421B08Rik

Create ppDesigner Target Files[edit]

  • Files in genome-miner:~/scratch/CA12kOct2016_ProbeDesign/MouseBrain
  • 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
 ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/opt/ppDesigner/src/ppDesigner.pl ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/jobFile_chr$indx.pl > ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/outputFile_chr$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 ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/ConvertToZeroGapProbe.pl ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/outputFile_chr$indx.txt > ~/scratch/TB12kOct2016_ProbeDesign/MouseBrain/173NewGenes/outputFile_0gap_chr$indx.txt &
 cat outputFile_0gap_chr*.txt > outputFile_0gap.txt
  • 2,389 probes
  • 1,435 exons
  • 172 genes

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

 /media/Home_Raid1/kunzhang/softwares/Novocraft/novocraft/novoalign -d ~/scratch/Genomes/mm10/novoalign/mm10_refMrna.ndx -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_mm10refMrna.out &
 /media/Home_Raid1/kunzhang/softwares/Novocraft/novocraft/novoalign -d ~/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

  • Output: outputFile_0gap_filtered.txt
    • 1,794 probes
    • 1,250 exons
    • 171 genes
  • Missing Trbc2

Combine Part 2 (173 New Genes) with Original 150 Genes[edit]

  • In GradZhangLab\TB12k_Oct2016\MouseBrain_V7
    • Add \173NewGenes_part2\outputFile_0gap_filtered.txt to outputFile_0gap_mod.txt -> outputFile_0gap_combined.txt
  • RevComp.pl

Add Barcode[edit]

 perl Probe2padlockFISSEQ_Oct2016_MouseBrain_6Cycle_V7.pl > padlockFile_0gap_MouseBrain_V7.txt

Lung Cancer Fusion Oncogenes[edit]

  • In GradZhangLab/TB12k_Oct2016/LungCancer_V8
  • Reference file: AmpliSeq_lung_fusion_panel_info.pdf
  • Sequence file provided by Dr. Zhang: Oncomine_Solid_Tumour_Fusion_REFERENCE_MAN0013544_RevA.FASTA
    • Identified junction positions by Blat-ing with UCSC genome browser and manually finding where the transcript aligns to two different genes
    • For imbalance genes identified the sequence that contained the exons contained in the name
  • Result: LungCancerPanel.xlsx

Fusion Transcripts[edit]

  • Files in GradZhangLab\TB12k_Oct2016\LungCancer_V8\Fusion_Transcripts
  • FusionCancerGene_ppDesigner.py in Dropbox\PythonScripts
  • Take LungCancerPanel_ForFusionCancerGene_ppDesigner.xlsx and run FusionCancerGene_ppDesigner.py
    • Uses Bio.SeqUtils MeltingTemp module to calculate Tm of RNA/DNA hybrid sequences
      • Tm calculated with salt concentrations of 1X SplintR Ligase Buffer
    • Searches for optimal annealing arms that are adjacent to junctions with length 18-28bp and Tm 60-65C
      • Optimal arm length = 22bp
  • Final output version: outputFile_1XSplintR_Tm_18-28bp_LungCancerFusionGenes.xlsx

Control and Imbalance Genes[edit]

  • In genomeMiner ~/scratch/TB12kOct2016_ProbeDesign/LungCancer/
  • Sequences: Ctrl_and_Imbalance.fa
    • Split into 13 separate FASTA files
  • Create jobFile.pl
  • Create targetFile.txt
    • Target Start and End are 25bp away from the beginning and end of sequences
 ./opt/ppDesigner/src/ppDesigner.pl ./jobFile.pl > ./outputFile.txt &
 perl ConvertToZeroGapProbe.pl outputFile.txt > outputFile_0gap.txt
 perl Probes2fasta.pl < outputFile_0gap.txt > outputFile_0gap.fa
 /media/Home_Raid1/kunzhang/softwares/Novocraft/novocraft/novoalign -d /media/12TB_ext/GenomeDB/HsGenome/1KG.ref/NovoAlign/human_g1k_v37 -f outputFile_0gap.fa -F FA -r ALL > outputFile_0gap_novoalign_human_g1k_v37.out &
 perl CleanupProbelist.pl
  • outputFile_0gap_filtered.txt
    • All 13 have probes
    • 37 probes total

Combine[edit]

  • Convert outputFile_1XSplintR_Tm_18-28bp_LungCancerFusionGenes.xlsx to outputFile.txt format by adding column that concatenates the two arms and removing header
    • outputFile_1XSplintR_Tm_18-28bp_LungCancerFusionGenes.txt
  • Add outputFile_1XSplintR_Tm_18-28bp_LungCancerFusionGenes.txt to outputFile_0gap_filtered.txt -> outputFile_0gap_combined.txt
    • For any genes with multiple probes in outputFile_0gap_filtered changed names by adding "_#" so each is uniquely named
  • RevComp.pl -> outputFile_0gap_combined_RevComp.txt

Add Barcodes[edit]

 perl Probe2padlockFISSEQ_Oct2016_LungCancer_5or6cycle_V8.pl > padlockFile_0gap_LungCancer_V8.txt
  • 4 fusion transcripts could not have padlock probes designed because the total length of arms > 51bp needed to fit in 195bp order
    • EZR-ROS1.E10R34.COSF1267
    • TPM3-ROS1.T7R35.COSF1273
    • LRIG3-ROS1.L16R35.COSF1269
    • GOPC-ROS1.G8R35.COSF1139
    • If we really need these they can be ordered individually from IDT
  • Not the same 6 cycle (4 'on' 2 'off') barcodes
    • Instead, this barcode can be fully decoded in 5 cycles and the 6th cycle is a redundant 'on' value
  • How to make this barcode?
    • Take 5 cycle (3 'on' 2'off') barcodes from decoding_linkers_5-cycles.txt
    • For each 5 cycle barcode add a '1', '2', or '3' to the end and alternate which one is added
    • Find the matching 6 cycle barcode in decoding_linkers_6-cycles_4-on.txt and use it to replace
  • New barcode file: decoding_linkers_5or6-cycles.txt
  • Each target has unique barcode

Combining into One Order[edit]

  • In GradZhangLab\TB12k_Oct2016
 perl padlock2oligo_order.pl > CA12k_Oct2016_Oligos2Order.txt
  • Total # of probes = 11,154 so copied LungCancer_V8 probes 8X plus a little remainder to bring up to 12k probes -> CA12k_Oct2016_Oligos2Order_FINAL.txt
  • Double checked probes for all 4 sets
    • Made sure they were 0 gap
    • Made sure they were reverse complementary to mRNA
    • Made sure barcodes were same as decoding probes (so that rolony would be reverse complementary to decoding probes)
Probeset Amplification # Probes # Barcodes # Genes
Human Brain V4 4,978 391 368
Mouse Brain V7 3,602 342 319
Mouse Embryo V6 2,469 251 220
Lung Cancer V8 105 105 85