Tuesday 01/07/14

10:00a – 10:15p

chromatin data collection

Finish O/N STORM on F04.
Transfer data from F03 and F04 to Monet
start analysis of F03 and F04 data
wash out stains for new BX-C regions: F05, G04 (ANTC), F03-F04-F05 combo, F05-A647 F06-A750.
image F05 regions (hope these work!)

Working on new model figure as slides for Ajaz — still haven’t finished this!
New Pc clustering — existing data pipeline for analyzing this is pretty good. Moved functional form of script into a Results code folder.

Substantial over expression — by quantifying localizations its 5x the wt levels. 100K localizations in both 750 and 647 per cell, whereas cells with no nuclear enrichment of 750 flag have 20K localizations of endogenous Ph.
issues with z-focus
- wt data was taken at lock target 0 for both 647 and 750. This seems not to align focuses– substantial new molecule regions evident in each channel not evident in other channel. Spatial distribution through nucleus seems to suggest a z issue. For example, the interior of the nucleus will be hollow and one channel will show a more filled middle and the other a more extended periphery.
- M data was taken at lock target 0 for 750, +60 for 647.
wt and mutant data good cluster overlap
images align better with drift correction turned off. Imaged based drift correction is just creating trouble.
caught and fixed a few little bugs in STORMrender

check probe alignment
wrote short script to BLAST against new internal database
legacy BLAST returns no hits. is it legacy BLAST that is strand specific not oligoArray? Or maybe I don’t have it configured right
BLAST+ works great, sequences all check out as desired. Only matlab readblastlocal doesn’t succesfully import the strand (should be Hits.HSPs.Strand, but this field doesn’t get created, even though I can see it on the Blast output).
updated my matlab BLAST function to default to display the name and sequence alignment of the top hit. would be awesome to get strand as well.