0501 Mon Meeting w/ Dr. Kreimer
- Analyses
- real motifs: overlapped WTs motifs: to be finished
- nonmotifs: overlapped: to be finished; MPRA output.
- Prediction: continue working on them
04.26 Tue Meeting w/ Dr. Kreimer
- redo analyses:
- F(all)=0 vs all, F(all)=1 vs all, between pertX
- nGained and nLost comparision
- overlapped WTs analyses
- predictions: ask Justin for the scripts and start working on it!
04.22, Fri, Meeting w/ Dr. Li
- manuscript:
- Fig1: add arrows; remove dashed boxes; DM names
- Fig2: QQ plot: gray x=y line; try -log10p; type I error as supplementary
- remove subsubsections
- check: under norm methods, gene variance changed?
- check: variance vs cell number in denSNE manuscript
04.18 Mon Meeting w/ Dr. Li
- manuscript
- finish results (effect of norm method & effect of nFeatures)
- schematic workflow
04.12 Tue Meeting w/ Dr. Kreimer
- activators/repressors:
- reverse the categorizing because log2FC = log2(WT/Pert).
- tfbs_func: 1&2 activators, 3&4 repressors
- read papers regarding MPRA and predictions
04.08, Fri, Meeting w/ Dr. Li
- manuscript:
- add toy examples for methods part
- QQ plot for type I error
- big picture: Best practices in differential variability analysis of single-cell gene expression data
- check: under SCT, gene variance changed?
- check: variance vs cell number in denSNE manuscript
04.05 Tue Meeting w/ Dr. Kreimer
- activators/repressors: add "filter" (F1-F4) criterias
- for pertX, wilcox test on:
- positive
- negative
- overall
- (also considering filters)
- add n_gained vs n_lost to slides
- discussion about joining the lab
04.01, Fri, Meeting w/ Dr. Li
- manuscript:
- focus on same cell number scenarios
- add toy examples for methods part
- big picture: evaluation
- Check: under SCT, gene variance changed?
- check: variance vs cell number in denSNE manuscript
03.29 Tue Meeting w/ Dr. Kreimer
- For the perturbed/not perturbed motifs, check if they have overlapped/not overlapped identical motifs
- what does each F mean?
- what does Filt mean?
- gained/lost motifs: activators or repressors?
- delta
$nMotif_{pert - WT}$ --> positive or negative?
03.25 Meeting w/ Dr. Li
- should keep an eye on Type I Errors
- Todo
- Try different cell number scenario:
- condition 1: 100cells/type
- condition 2: 100-1000cells/type
- Update manuscripts
- Try different cell number scenario:
03.24 Meeting w/ Dr. Kreimer
- Fisher's test for subgroups?
- crosstable, chi-square?
- add raw numbers of motifs
- Wilcox on F=0 motifs