Post candidacy: all the crap in lab

It is annoying to argue anything with my PI.

Regardless of what I argue, how I argue, what data I present, he is always glued to his stubborn opinion.

Last week, he forced a change in imaging medium on the ground that the CO2 supply was changed thus the imaging medium needs to be changed accordingly. Despite all the arguments sounds logical, the fact remains that this might be incompatible with my experiments due to my specific procedure that took place prior to the imaging.

Today I was able to demonstrate this in live-cell imaging.

Yet he came back with all these arguments that these two cells are different and I shall not change back to my old reagent, even though the old reagent consistently makes my cells respond to stimulus and have a stable phenotype.

I could never convince him that the cells in this cell line are heterogeneous, which they are, and they are known to be, and I have seen enough at my hand.

I don't know what to do when he gave all this shit.

He always use "PIs are expected to be suspicious and challenge their trainee" as an excuse, while being an asshole around, trashing my work, while doing nothing to help.

His instinct is not to think, but fall on his "common sense", which only God knows where are based on and presumably has something to do with Jackson County, MI, and then start to argue against you, instead of support your experiments.

One of the postdoc once said jerks breed like rabbits. I think they are more likely to breed like mice, and definitely imbreed in academia, so I guess nothing is new.

Post candidacy: When you make a mistake

When I made a mistake, my PI never let it go. It would be mentioned again and again.

When he made a mistake, he shrugged it off, saying everyone made mistakes and should move on.

The part that he admitted him made a mistake was not a part of everyday life, but the part he accused me of doing sth wrong is the part of everyday life.

100 ways for experiments to go wrong in a biology lab

I start this log because there are so many things happened in my lab that I am not even sure I should cry or laugh. Thus, I decided to start writing down stories that happened in this lab so that the readers can judge themselves.

Chapter Li, Winter, 2018

I was trying to reproduce an immunostaining image in the literature.

My PI asked me to try some random dye to stain the nuclear of a cell, rather than using DAPI, a well-established reagent. I asked why I could not use DAPI, he commented on he didn't wish to waste money on another laser line.

I quietly followed his protocol. Initially, staining nuclear with ToPro3, which fluorescence could be collected with 647 channel, worked.

Then I started to see weird patterns on my images, and eventually, being able to collect no good data.

I mentioned this problem to my PI. He said I must have done something wrong, and I should simply repeat everything again with his watchful eyes.

I tried. Then the weird pattern on the image attracted his attention.

It turned out that the filter for the fluorescence collection was burnt. Thus it was normal that I could not get anything useful.

Chapter Li, Winter, 2019

A more detailed version can be found here.  In short, my PI asked me to purify a protein. I failed 3 times, and every time he accused me of having bad hands.

It later turned out to be he gave me the wrong protocol. He mixed up the formula between elution buffer and wash buffer. For the whole time, I had been, under his instruction and supervision, using elution buffer to wash my beads for purification.

Chapter Milda, Spring, 2019.

Milda was trying to re-produce a published result, which is YAP, a protein, is localized in the cell nuclear when the cells are on a stiff matrix. Yet she saw no signal in her immunofluorescence image.

My PI started with speculating Mida did something wrong, and then speculated the antibody might not be the best. But nothing seemed to be the cause. He then asked Dr. Rodionov, a person who has decades of experience in immunofluorescence and extremely nice, to help Milda to troubleshoot.

Dr. Rodionov could not find any apparent issue. However, he suggested that Milda could use his more expensive formaldehyde, rather than the cheap substitute in my lab.

And then everything worked. There was bright immunofluorescence of YAP, overlapping with Milda's DAPI stain for the cell nucleus (DAPI).

Chapter Prem, Summer, 2019

Prem is interested in regulated secretion. He is using fluorescent protein-tagged insulin to gauge the process of insulin secretion, and other proteins that may be involved in it.

In the literature, people used C-terminus tagged yellow fluorescent protein (mVenus, a bright red-shifted variant of GFP) to achieve this goal.

In his case, he used a C-peptide tagged human insulin. Previous research suggested that this is a better fluorescent protein tag than the C-terminus tagged. They are more likely to be folded correctly.

Previous in house data suggests this C-peptide tagged insulin goes to granule like structure and can be trafficked in the MIN6 cells, a pancreatic beta cell line with mouse origin. This insulin can be secreted out of MIN6 cells when the cells are stimulated with glucose, like real beta cells. This was shown in both literature using biochemical data and in-house with live-cell imaging.

Prem wished to use total internal reflection fluorescence microscopy to gauge insulin secretion. Thus, he tried to switch the fluorescent protein tag from GFP to pHluorin. pHluorin is a pH-sensitive variant of GFP. In the acidic environment, pHluorin is dim; in neutral pH, the fluorescence becomes brighter.  Secretory granules are acidic, while the extracellular medium has a neutral pH. Thus, it was believed that when the MIN6 cells are stimulated with insulin, he should be able to see lots of bright puffs that signals insulin secretion.

However, in the real experiment, he saw nothing.

Later research suggests that pHluorin might go through a different post-translational process (i.e. different glycosylation). This may be the reason why he cannot see anything in his experiment.

Chapter Li, Summer 2019

I have been trying to capture insulin receptor A, an isoform of the insulin receptor, enrichment in primary cilium in MIN6 cells. I was able to capture ciliary enrichment of human insulin receptor A with interdomain GFP tag using immunofluorescence.

However, I was unable to achieve the same thing in my live-cell imaging.

My PI thinks this is due to the GFP is too dim so there isn't enough signal for me to capture. Thus, he forced me to switch the GFP tag into a more brighter fluorescent protein, called mNeonGreen.

Yet still. I see no IR-A-mNeonGreen in the primary cilium.

Then I tried to immunostain the cells to see whether IR-A-mNeonGreen goes into the primary cilium anyway. To my surprise, I cannot see the receptor's ciliary localization.

Then I ran a western blot, comparing IR-A-GFP and IR-A-mNeonGreen expression in cells. Oddly, despite mNeonGreen is only 3 amino acids difference from GFP, there is a significant molecular weight shift in the mNeonGreen tagged receptor.

I guess this is another "post-translational modification messed up my experiment"!

I hate all these fluorescence protein manufacture not to put these on the warning.

What a waste of time, my youth and energy.

Chapter Li, Winter 2019

I was trying to repeat some experiments I did in the past. This is an experiment to test whether our design of photo-activatable insulin receptor (optoIR) can be activated by light. If our design works, we are expected to see light-induced receptor phosphorylation.

We were able to see light-induced receptor phosphorylation for a couple of times. But later we cannot repeat the experiment after we start to illuminate our sample from the top through the lid, rather than from the side.

We later found out that we cannot see light-induced phosphorylation because of water droplet condensation on the top of the lid. The condensation blocks the light thus we see no light-induced receptor phosphorylation.

Chapter Li, Winter 2020

I was trying to repeat an experiment I did before. Previously, when I stimulate my cells with this stimulus, I was able to see fluorescence change from the cells that express a fluorescence-based biosensor (fluorescent protein) on our microscope. The experiment outcome has always been repeatable.

But somehow I didn't see this happen in a January day.

It turned out that my PI changed the recipe of our imaging medium.

Everything worked again when I changed the medium back.

Post candidacy: the tale of protein purification

Last night I shared my experience in protein purification with my roommate, who is specialized in studying protein. She laughed so hard that I think I should write down my story, which took place this February.

My PI asked me to purify a protein, a FRET sensor with HIS tag at the N teminus. He gave me a protocol out of his head, claiming this was what he had been doing for a decade and this is the protocol that worked for the previous graduate student, Abhi.

He asked me to start by preparing the bacteria soup in the volume of 100ml. Yet I can see no fluorescence in the tube after elution. He accused me of having bad hands so that I must have lost all the proteins during purification.

He then asked me to prepare the E.coli soup in 1 litre so that I could have more room to make mistake yet still yield purified protein. The bacteria soup and all these reducing agents smelled so terribly I had to wear a plague mask, but even that I still suffered from a terrible headache induced by the stench. Yet nothing worked.

He then accused me of not following the protocol religiously and asked me to have him watch me when I was doing my experiment.

I did it again. I showed him the bright yellow fluorescence (from the YFP in the FRET sensor) in the bacteria pallet yet nothing yield in my elution solution.

It was the day of recruitment. I remembered potential candidates walked in the lab, and I was on my plague mask. I had to apologize for the stench.

Later after dinner with candidates, he approached me and told me that he made a mistake. He mixed up the formula between elution buffer and wash buffer.

For the whole time, I had been, under his instruction and supervision, using elution buffer to wash my beads for purification.

I guess it is always a great story to tell.

PhD post candidacy: mourning

It has been two month since the death of my beloved Sangria. Legally she is my property. Traditionally people don’t mourn for the their dead pet publicly. Yet I find it hard to get over the grief. She was one of the best things happened in my PhD years. She helped me calming my nerve, release my stress, and give me an excuse to rest at home. Without her, I am drown in work, and I feel my emotion and mental status are at the brink of collapse.

Yet my work keeps me so busy that I barely get a minute off to write a proper eulogy for her.  Every falling flowers keeps reminding of the ending of her life. I am not sure whether it is the pollen or her keeps me crying.

I could never commemorate her too much.

鷓鴣天Sangria
不解街邊貓沈思，未讀國風嫌惡詞。灰鬢不度香腮雪，偷嗅鏡中影是誰。 柳絮嫁，杜宇離，鶯愁蝶倦晚芳時。只怕惜花心易碎，青冢不見荼蘼枝。

Post candidacy: PhD is a scam

I realized something funny when I was talking to my boyfriend.

About two years ago, I was preparing for a journal club. In that paper, the authors used both mouse and rat models. I asked my colleague, a person that hold PhD from a German institute, what the difference is between mouse and rat. He shrugged, and replied he wasn’t sure. He then commented “I think there is a difference, but I don’t know. It’s difficult to tell since English is not my mother tongue “.

Post candidacy: from cheer to woe