Mrinal Joshi / Neurogenomics Lab, Imperial College London
A day at
the bench
I edit genes in human cells and look for what changed. Try a few steps from my work, with weeks of waiting squeezed into one day. You'll write a tag into DNA and pipette, then look at cells I grew and read a simulated qPCR run.
Drag across the dish. The culture grows wherever you touch it.
Write a tag
into a gene
I can't pick out a particular protein by looking down a microscope. To follow one, I add a short sequence to the end of its gene. The cell makes the protein with a small tag attached, like a luggage label. Later I can collect it using an antibody that binds the tag.
Move a few microlitres
You could fit a thousand microlitres into one millilitre. I move these small volumes with a pipette that has two stops on its plunger. Push to the first stop before you draw up slowly. Over the well, push to the second stop to empty it. Let go too fast on the way up and you can pull in air.
- 01 Set the volume
- 02 Draw up
- 03 Find the well
- 04 Push it out
I use a transfection mix to help DNA enter the cells. You'll see some of my cells under the microscope at 14:20.
Look for the glow
A couple of days after pipetting, I put the plate under a microscope. You can see the cells in white light. Switch to blue light and look for green fluorescent protein, or GFP, to check DNA delivery and expression. These are my pictures of HCT116 cells, a human cell line used in research.
Glow gene only
I added a plasmid encoding green fluorescent protein, or GFP. Green cells have taken up and expressed the DNA. I use this positive control to check delivery.
Find where it sits
I study proteins that bind DNA and help control gene activity. With TIP-seq, I can map where they bind. Scroll through the four steps to follow the tagged protein.
- 01
Grab it
I add an antibody that binds the tag on my protein, then wash away what hasn't bound.
- 02
Mark the spot
I bring in an enzyme through the antibody. It cuts nearby DNA and inserts a short sequence called a T7 promoter.
- 03
Copy it
I use T7 RNA polymerase to make RNA copies of the nearby DNA, starting at those promoters. Then I convert the RNA back into DNA for sequencing.
- 04
Read it
I sequence the copies and match them to the genome. Clusters of reads mark places where the protein may have bound. I compare them with controls before calling a site.
Call the round
I use qPCR to estimate how much more copying a DNA library needs before sequencing. You'll watch a dye glow as DNA builds up. In this game, call the first round that reaches a quarter of each curve's rise. In the lab, I choose the cycle count using the library prep method and quality checks.
- 09:10Tag writtenNot done yet
- 11:00PipettingNot done yet
- 14:20Cells foundNot done yet
- 16:40Cycles calledNot done yet
Day average--
Nothing in the book yet. Scroll back up and get your hands dirty.
AI at the bench
I use AI for coding and working through papers. I'll write here about trying it at the bench, including the mistakes I catch along the way.
First posts soonRead the notesMrinal Joshi
I'm a research assistant in the Neurogenomics Lab at Imperial College London. I work on ways to tag proteins in human cells and map where they sit on DNA. You've tried a few of those steps here. In the lab, I spend weeks growing cells and checking the results.
- NowResearch assistant, Neurogenomics Lab, Imperial College London
- MScRegenerative medicine and bioinformatics, Imperial. Distinction.
- BScMolecular genetics, King's College London. First class.