Matt:LabNotes/2015-5-14: Difference between revisions
Jump to navigation
Jump to search
>Mzcai |
>Mzcai mNo edit summary |
||
Line 41: | Line 41: | ||
**Is this due to BA sequencing by Smart-Seq bias? | **Is this due to BA sequencing by Smart-Seq bias? | ||
[[File:DARTFISHgNorm_vs_HBRR.png|450px]] | [[File:DARTFISHgNorm_vs_HBRR.png|450px]] | ||
===DARTFISH vs BAs and HBRR=== | |||
*DARTFISH counts normalized by cDNA counts | |||
**cDNA counts normalized by HBRR RNA-Seq FPKM (divided) | |||
**All correlations improved equally approximately | |||
[[File:DARTFISHcNorm_vs_BA8.png|450px]] | |||
[[File:DARTFISHcNorm_vs_BA10.png|450px]] | |||
[[File:DARTFISHcNorm_vs_BA17.png|450px]] | |||
[[File:DARTFISHcNorm_vs_BA21.png|450px]] | |||
[[File:DARTFISHcNorm_vs_BA22.png|450px]] | |||
[[File:DARTFISHcNorm_vs_BA41.png|450px]] | |||
*HBRR | |||
[[File:DARTFISHcNorm_vs_HBRR.png|450px]] |
Revision as of 21:50, 18 May 2015
Regression Analysis of CA12kNov2014 V4 probeset + 100x suppressor
- Files in Dropbox/GradZhangLab/CA12k_Nov2014/V4_CaptureAnalysis/20150514/
- DARTFISH counts from Summary2_S1_V4_Supp_2015-05-06_Try1.txt
- 18 Positions of V4 in BA8 using first generation of 100x suppressor oligos for CA12kNov2014 probeset
- HBRR FPKM values from: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE49712
- File: GSE49712_gene_FPKM.txt.gz
- Experiment was sequencing UHRR (Sample A) and HBRR (Sample B) and did 5 replicates of each
- I averaged the FPKM of the 5 Sample B replicates
- File: GSE49712_gene_FPKM.txt.gz
- Brain Tissue RNA-Seq FPKM Data:
/media/LTS_33T/RL_LTS33T/201404_201405_7Samples_BulkNucleiBatch1-20140623_Expt146/STAR/
- Files moved to Dropbox/GradZhangLab/CA12k_Nov2014/V4_CaptureAnalysis/
- eg. RL-BA8-sec9-t-N701-15May14_S1_mapped_genes.fpkm_tracking.txt
- BA8, BA10, BA17, BA21, BA22, BA41
Normalize in vitro Capture and check with RNA-Seq
- Captured cDNA from HBRR
- Normalize with gDNA counts
- Compare to HBRR FPKM values
- Better than comparing with BA8 (as expected)
- Some correlation also expected since HBRR is from all areas of brain and many samples
DARTFISH vs BAs and HBRR
- DARTFISH counts normalized by gDNA counts
- Despite DARTFISH being from BA8 it is the worst correlation...
File:DARTFISHgNorm vs BA8.png File:DARTFISHgNorm vs BA10.png File:DARTFISHgNorm vs BA17.png File:DARTFISHgNorm vs BA21.png File:DARTFISHgNorm vs BA22.png File:DARTFISHgNorm vs BA41.png
- HBRR has best correlation
- Is this due to BA sequencing by Smart-Seq bias?
File:DARTFISHgNorm vs HBRR.png
DARTFISH vs BAs and HBRR
- DARTFISH counts normalized by cDNA counts
- cDNA counts normalized by HBRR RNA-Seq FPKM (divided)
- All correlations improved equally approximately
File:DARTFISHcNorm vs BA8.png File:DARTFISHcNorm vs BA10.png File:DARTFISHcNorm vs BA17.png File:DARTFISHcNorm vs BA21.png File:DARTFISHcNorm vs BA22.png File:DARTFISHcNorm vs BA41.png
- HBRR