10:00 A – 7:00P, 8:30P-10:00P
STORM
- Finish O/N run of PhM-flag-488, Ph-647
- take bead images
- transfer data to ProRAID
- Set up analysis to run on Cajal
Embryo in situs
- wash out probes
- check staining: All embryos washed off!
- 1 tiny section remaining on coverslip 1 (HCl treated, primary + secondary) No sections found on previously excellent coverslip 2. No engrailed expressing cell re-located.
- complete and horrible failure yet again at En protein + DNA region doubles…
HoxD
- HoxD probe finding finally complete.
- 30K probes (skipping the interval regions for the most centromeric loci, which are respectively 32.5K and 12K probesets)
- All regions have over 100 probes (min size = 120 probes, median = 583).
Literature
- Reviewing 2013 Hi-C literature
- read Cavalli review and commentary
- good general multiC review by Dekker in Nature Genetics.
- several structural models out: only the simplest like binding sites in region cause like clusters looks reasonable. Infering 3D structures of what might be random polymers from interaction data is BS. What if within a domain, all individual contacts look different.
- Several software pipelines available for analyzing Hi-C data, none seem to be easy to run and take raw read or paired-region data (raw data on Corces Hi-C paper is paired map locations, not yet mapped back).
- Li/Corces online data claims it uses the Yaffe bias correction. Recent Harvard Stats code claims to accelerate Yaffe method >1000 times.
- could request un-normalized data and for annotated ‘region boundaries’?
presentation prep
- going through Hi-C data
- added visualization of color regions with Hi-C data + indicator color tracks.
oligoPaint STORM
- analyzing BX-C sample 1.
- Nice bright dots, dense cells.
- Not much switching after bleaching. recording 32K frames.
- Beads didn’t stick except few clumps. Need to make new beads. Give beads longer to settle.