Matt:LabNotes/2016-4-8
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)
- 42 genes left: "42genes.txt"
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 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
- 198 Unique Ensembl Gene IDs and 198 Unique HGNC Symbols
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