Hello friend,

“Why on earth would you search for a drug to reduce glucose?”, you may ask.

Well, well, well, young disciple…
The reason is very simple and yet very, very complex:
to justify eating ice-cream, for the sake of science.
For that, young warlock, is the purpose of life. Anyway..

“So, what’s Berberine then?” Berberine is a plant-extract that has been shown in both human and animal trials to be as effective as pharmaceuticals for treating type-II diabetes. Unfortunately, all the clinical trials I could find only showed how it reduced biomarkers of diabetes, such as fasting blood glucose and HbA1c, but none showed the actual trend after ingesting a meal. This, you may have guessed is important on the days you want to have the upsides of pizza, with none of the downsides.

So I decided to try it home instead. The result? It exceeded my expectations.

How the experiment went

• Every day at the same time I took 150g of cooked rice (that is about 45g of carbohydrates).

• To measure my glucose, I used a continuous glucose monitoring (CGM) and analyzed my data using glyco in Python.

• Most trials use a dose of 0.5g of Berberine, 3 times a day. Instead I took a dose of 2g in one meal. Go big or go home, right?

• I either took the Berberine or a placebo (used Berberine-500 from Thorne Research). And I either took them 15min before or right after a meal. I sat down for 2hours after the meal, and avoided exercise and food before the rice meal.

Results: how Berberine exceeded my expectations

As you can see in Fig. 1, you don’t need statistics to see how big of an impact it has on glucose. When taken 15min before the meal, Berberine largely reduced my peak glucose, but also the area under the curve of glucose. The effect was so large I could guess that I did not have placebo that time.

This should not be surprising given the large body of evidence supporting Berberine, including mechanistic evidence about how it indirectly activates AMPK [Brusq et Al. 2006]. But seeing it in real-time using the CGM left me stunned. I looked at the data, looked back at the empty bowl, looked back at the data, I could not believe it.

I have to admit, however, that I stopped the experiment early. While the original plan was to do this for 10 days, ingesting 150g of rice made both my heart-rate and blood pressure go up.

For this reason, I was worried the effect might have been due to something else (you never know ¯\_(ツ)_/¯). So I tried again, this time with a snack bar that has about 15g of carbohydrates. And once again, the drug did the job (Fig. 2 below).

As skeptical as I was at the beginning of this experiment, these results were not only convincing, but impressive.
I am now officially a Ber-believer. Yes, yes, I just made that up.

Discussion: Why I don’t use Berberine daily

After seeing the dramatic effect on glucose, I still do not to use it everyday. Instead, I would probably use it when exposed to high amounts of glucose in parties or other uncontrolled high-carb high-GI scenarios.

Other than the extra 2$/meal it would cost, there are two more reasons why I would not use Berberine everyday:

1. When I take Berberine, my glucose drops (Fig. 3 below). This drop is so abrupt that I feel tired afterwards. This feeling was worse when consumed in a fasted state.

2. In the world of supplements, there is no such a thing as a free lunch, and the risks of taking a supplement must always be weighed against the risk of not taking it. And in this scenario, I think staying relatively low-carb is a better strategy.

3. Don’t try using Berberine as a powder in ice-cream. The taste is not worth it (and believe me, my taste is as stoic as it gets) and it will turn your teeth yellow (unless that’s the style you’re going for).

For those of you who skip directly to the end of this article

Here’s the summary: Berberine largely reduced my glucose levels after a high-carb meal. It is now an essential supplement in my ice-cream emergency cupboard, but I would rather avoid it in regular days and just stick to lowering carbohydrates in general.

Thank you for reading, if you liked this post, please give it some claps, it’s easy but it would mean a lot to me.

How we use a central protobuf schema to make our data pipeline less chaotic

If I told you one of the most popular TV Shows in Sweden is about following the moose as they travel, for 21 days, you might say:
 — Stop it with the stereotypes!
or :
— Ismail, go home and stop experimenting with psychedelics

I promise, this TV show is the real deal and it’s only 21 days long.

The TV Show “den stora älg vandringen” (which translates to “the great moose walk”) is a type of TV Show known as slow TV. This trend started in the neighboring lovely country of Norway (at NRK).

This TV show is 21 days long and it was streamed live on SVTs (Swedish Television’s) streaming service SVT Play.

So why collect data then?
It turns out filming the moose is not the only thing that SVT does and SVT Play is just one of the services at SVT. These services range from streaming to delivering news in addition to games and services for children and more. The point is, those apps make different types of content available (news articles, movies, TV series etc.) on different platforms including web, iOS, Android and smart TVs.

So a little more than two years ago, a project for data collection at SVT got started. This was an effort to aid decision making in the multiple products of SVT (that used to rely on third-party services such as google analytics).

Of course, the data is not what makes or breaks the products, nor the shows, movies or news articles. It is there to verify assumptions, to understand the users better and to support the decision making.

As one can see, the diversity of SVTs services and the platforms on which they run adds multiple challenges to the collection of data.

Too long; didn’t read?

If you prefer watching this as a video feel free to head over here.

While there are many things to talk about when it comes to the collection and analysis of data, this post will focus on:
— A very brief intro to how we collect data at SVT.
— The chaos that can result from the changes in the business rules.
—How we use a central schema (with Protobuf) to define the data we collect and propagate this definition in the whole pipeline (from collection to storage).
 — Some of the issues we encountered with that choice.

A Brief on how we collect data at SVT

Let’s say you’re a news editor, and you want to see how your articles perform. How can we collect data about that?

For example, you would want to know: how many users read an article? We cannot know for sure if a user actually reads the article, but what we can do is estimate it from what we can collect such as: a mouse click on the article or a mouse scroll inside the article.

These things that we can collect are what we call “Events”. Other example events for other products can be:

• A video start or a video pause in the video player.

• A search query in a search bar.

• Or other events for collecting video player telemetry information.

An event is an action that happened on the client application side, for which the corresponding information can be collected (for example: a click or a video start)

These events are just observations, and from them we should be able to either calculate or approximate our metrics (an example of a metric that cannot be calculated exactly is the percentage watched of a video because we simply do not know if the user is actually watching)

So we send those events. We ingest them into our pipeline (more details later). We store them in a table. We process and analyze them and then we present them to the users of our solution.

So the event will go through:

• collection: on the user-facing application side. For this part, in our case, we built SDKs (mainly for iOS, Android, Web and smart TVs) that will collect events such as clicks and send them to an API, which brings us to the next part.

• ingestion: when the API gets the previous event it publishes it to a message queue. In our case, we are currently using Pub/Sub on the google cloud platform (GCP). We have creatively named the previous API: the event API.

• storage: a service gets the previous event from Pub/Sub, checks the validity of its content, then stores it in a table (a BigQuery table in our case). We have called this service judge-judi.

• analysis: mostly just a bunch of ETL

• presentation: at this stage, we can show a dashboard to the right person with the right metric.

For the purpose of this post, we will only focus on the collect — ingest — store parts of the previous pipeline.

To keep things simple, we can assume that we send one event at a time from the client application (In reality, those events are gathered by the client application and sent together in a sequence of events)

So, with all this in mind, where can problems happen?

The need for ONE (versioned) definition of the event data

From what we’ve talked about in the previous section, here’s what the life of an event looks like from collection to storage:
 — First, we send the event from an app (through our SDK) to the Event API.
 — The Event API then adds it to the queue for it to be processed later.
 — And then on the right, is an accurate picture of judge-judi (the service described in the ingestion part above) performing its duty of pulling and storing valid events in the table.

The first problem is that all parts of the pipeline, from collection to storage, have to agree on what the definition of an event is.

The first problem: consistency

Let’s assume that the event looks something like:

{
    'user_id':   '00000000',
    'content_type': 'video',
    'event_type': 'video_start',    
}

Given the event above. What if one of our products ends up sending an event by naming the type eventType instead of event_type?

{
    'user_id':   '00000000',
    'content_type': 'video',
    'eventType': 'video_start',    
}

When this event reaches the validation stage at judge-judi. The service is expecting a field event_type it will just conclude that the field is not there.

So the first problem is having all parts of the pipeline agree on the definition of the data.

We also have a multitude of products in SVT (svtplay, news etc.) and different platforms (web, iOS, Android etc.) it would be hard to make sure that the SDKs are always consistent.

The second problem: Murphy’s law of data engineering

Which states that: “If the data schema can change, the data schema will change.” Now, Murphy did not necessarily say that.
Well, he never actually did, but I’m sure he would agree.

The business rules and goals change more often than teenagers’ relationships. We should expect, as a result, that the definition of the event will change. And it will change often. New fields will be added, old fields will be removed, others merged together, and the list goes on and on.

These modifications put schemas, and therefore the definition of an event, in a state of perpetual change. Each change will require adjustment to the pipeline. We do not want these recurring changes in the definition of the data to break the whole (or part of) the pipeline. Ideally, we would want to preserve backward and forward compatibility whenever possible.

And then problems specific to JSON

• The type of a field in JSON is unclear and errors can happen. It would be preferable if we can provide guarantees of what the event structure to be expected is.

• Because we have different platforms, we are using different programming languages. This means that the JSON body can differ depending on the programming language and libraries used.

• Long serialization overhead on the Event API’s side with the use of JSON. Although I currently do not have the “data” to prove it, it is easy to see that the schema is sent with every event.

But since we have control over the applications and have control over the schema definition how can we make use of that?

The solution: centralizing the definition of an event with Protobuf

One way of tackling the issues above is to do the following:

• One repository where the definition of an event, which we will call the Event Model, is kept (we just keep it in a repository on Gitlab).

• This event model is versioned.

• Every version of the event model is then propagated across all the elements of the pipeline. For example, the web SDK will have access to an event model library and every new version in the event model repository will correspond to a new version in the library. This version should be the same for all platforms.

• We will preferably use a language-neutral definition for the event model’s definition that can enable us to serialize an event on the SDKs and then deserialize it on the side of judge-judi.

• Preferably adding or removing fields from the event model should not break backward/forward compatibility.

To make a choice we tried both Avro and Protobuf. While there are many choices when it comes to defining the schema, to have a definition that supports all of the languages and provides a “good enough” serialization: Protobuf was the tool of choice.

What is protobuf and how does it work?

Protocol buffers are a language-neutral, platform-neutral, extensible way of serializing structured data. At least that’s what they’re supposed to be.

We will not go too deeply in how protocol buffers work but here are the basics.

Let’s say Alice wants to send a letter message to Bob that would look like this in JSON:

{
  'firstName': 'Alice',
  'lastName': 'Alice',
  'email': 'alice@alice.alice',
  'content': 'Hi Bob, it's Alice'
}

To do so, in protobuf, Alice and Bob would have to agree on a message schema. This schema can look something like this:

message Letter{
  required string firstName = 1;
  required string lastName = 2;
  optional string email = 3;
  required string content = 4;
}

This letter schema will be stored in a .proto file.

Alice and Bob can then convert this message to C++ or Javascript code or whatever language they’re into using the proto compiler.

Using the generated code Alice will serialize the message. For example, if Alice is using C++ her code might look something like this:

Letter letter;
letter.set_first_name("Alice");

Alice can now send the letter to Bob. A series of bytes are sent over the network. Bob knows the Letter schema, Bob can deserialize the letter message.

So how can we use it for the solution above?

Step 1 — In our case, the event model repository hosts the .proto file.

Step 2 — This repository propagates code with the same for the rest of the pipeline through pushing libraries. This repository also creates the table on BigQuery (as we will mention in the next section, this part has changed).

Step 3 — The service judge-judi will also fetch the latest Go version of the event model because it will need it later. And, just like Alice, through the SDKs we can fetch the latest version of this event model on the client apps to serialize the events that we will send. The app will serialize the event and send it to the API.

Step 4 — We collect this event as it is, without deserializing it, and we publish it to the queue on Pub/Sub.

Step 5 — The service judge-judi pulls the events and deserializes them.

It’s all good now, you drop the mic, walk out of the room, head high…until…stuff happens. You realize that the reality is as far from your expectations as your tongue is from touching your nose… So close, and yet so far (don’t try that, you’ll look ridiculous)

Expectations vs. reality

The myth of language agnosticism

When the mechanism of protocol buffers is supposed to be language-agnostic and platform agnostic. In practice, this does not seem to be the case.

For example, some platforms had encoding that was not consistent with the deserializer. We use the gogofaster compiler for Go which is a third party implementation that leverages type-specific marshaling code for extra performance. But we use it mainly to provide more canonical Go structures, less typing and to not end up with the infamous xxx fields.

The myth of backward and forward compatibility

Let’s say a field gets added, and a field gets removed from the schema. If the service judge-judi is using the latest version of the event model, it will still be able to deserialize the data with no problem even if the client application does not update to the latest version. But this is only true for simple changes. For example, this would not hold if we change the nesting of some fields.

Moreover, although we avoided talking about the analysis that comes after the storage part earlier, a new event model would mean that all the jobs and tables that depend on it will change. For this reason, we separated the version of the event model from the version of the event table on BigQuery so that the table would change less often. We then provide a mapping on judge-judi from the event model to the table schema before storing the events.

Thus, we identify mainly three types of changes in the event model’s schema and act accordingly:
 — Non-breaking changes: that happen regularly. These are the changes that do not break backward compatibility. This can be the deletion or addition of fields in the protobuf schema.
 — Breaking changes that do not require a change of the BigQuery table: that rarely happen, and should be minimized. When these happen judge-judi just changes the mapping between the event and the table.
 — Breaking changes that require a change of the BigQuery table: very rare and have yet to happen but they could (thus they will). Since those are rare and we cannot know what these will be like we’ll tackle them on a case by case basis.

Conclusion

The reality of data collection and analysis is very different from the mathematical fantasies of well-defined axioms and controllable assertions.

Coming from a scientific background, we have a tendency to want to keep the pipeline clean. And that’s good. However, chaos in data engineering is not just real but probably unavoidable. And it is one of the reasons why we should try to iterate slowly, and add tests and data quality checks when possible. It is also why keeping a central definition of the data in one place can be very helpful.

Protocol buffers are one way of defining a data model that seems to improve serialization overhead and be almost language agnostic. If you already know what the data schema should look like, protobuf is good at managing additions and deletions of fields, although one has to be careful as always. More complex changes will require more thoughtfulness.

Hope you found something useful reading this… Have a beautiful day!

If you have any comments or suggestions would love to hear from you.

Leave a comment

Make your data pipeline less chaotic by one versioned definition of data was originally published in The SVT Tech Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.

Hi,
I’m Ismail and I have a confession to make…
I take sauna baths… a lot of sauna baths.
Some days, I also have a sensor stuck to my left arm.
Other days I prick my fingers.

What do these things have in common?
Well, apart from them sounding like a form of self-torture, they’re also a form of self-experimentation with the same goal in mind: keeping blood sugar levels under control.

But why care about glucose?
To keep it short, if you want to perform better physically and mentally or just want to feel great, regulating blood glucose is among the 20% of things to focus on that give you the 80% of the outcome. For example, a study published in the Diabetes Care Journal found that during acute hypoglycemia (low blood glucose): speed of information processing, working memory and some aspects of attention were impaired. On the other hand, during hyperglycemia (high blood glucose) the subjects were significantly more dysphoric, had reduced energetic arousal, and had an increase in both sadness and anxiety. So, if we truly care either about feeling good, limiting negative feelings and increasing both mental and physical performance managing glucose swings seems to be not just impactful but necessary.

So who’s this post for?
I strongly believe that the protocols mentioned in this post are worth a closer look as upgrades in the toolbox of a lot of people, from elite athletes to those of us who want to perform better, or just feel better, during most days.
It could also be useful if you want to perform beyond your current level. For example, improving glucose variability can help athletes break through their endurance threshold (a lot of which would dramatically call the “wall”) by enabling them to use less glycogen from the muscles.

The goal of this post is to go through strategies that go beyond just diet, sleep and exercise to do that.

A word of caution:

Heat can be seriously dangerous
I am not a doctor nor an expert of any kind so please don’t be dumb and consult with a health care professional before doing anything described in this post (full disclaimer at the bottom of the page).
This post will talk about the benefits of sauna bathing, it is good to know beforehand that heat can be seriously dangerous and people can die from excessive heat. Sauna use during pregnancy may carry some health risks to a developing fetus. Please take the disclaimer at the bottom of this page seriously.

Why go beyond diet sleep and exercise?

You just cannot beat diet, sleep and exercise.
Although the goal of this post is to go through uncommon methods for controlling blood glucose levels, we cannot do better than: eating well, sleeping well, and exercising regularly.

While it is clear how diet and exercise can impact glucose. Sleep still seems a little underrated. In my own case, a lack of sleep (especially for extended periods of time) seems to correlate with high morning blood glucose. Just like the example in the picture below, this seems to hold even when my diet is all under control:

Running a controlled experiment on this would not be a good idea, but this result is far from surprising. In a study that investigated the effect of sleep debt on metabolic and endocrine functions there was a clear consistent impairment of how the participants tolerate glucose when they were in sleep debt (the rate at which sugar was cleared, after injection, was about 40% slower in sleep-debt than in the sleep-recovery).

But what if we keep those above variables (meaning sleep, diet and exercise) under control and you want to go a step further? Even if those last variables are the 20% that give you the 80% of the outcome. Our goal here is to go beyond that and reach the final 99%.
Plus, some days you just want to have cake, other days you just have to wake up earlier. What can you do then to feel good and manage the day? How can you have your cake and eat it too?

So let’s get started. And feel free to tell me which tools you use in the comments.

Leave a comment

1) Take a sauna

In general, heat is uncomfortable. After all, cooling fans disappear from supermarkets during hot seasons and people take refuge in air conditioned shops.

That said, it appears that increasing your core temperature for short bursts is good for general health. More than just aiding with glucose control, heat stress can also drastically aid performance, increase muscular hypertrophy, endurance and cognition.

When it comes to saunas I might be a little biased. I usually take enough saunas during the week to get comments such as: watch out, you’ll end up like a dry sponge at some point.
At some point though, I tried wearing a continuous glucose monitor during sauna sessions. I noticed the following:
 — my glucose levels seem to rise during the sauna usage.
 — after the session is done my glucose levels stabilize to a lower value than they were before.

Note that it is possible that this spike is caused by a problem in the sensor’s measurement due to the heat or dehydration. Even if that were true, the levels of glucose seem to stabilize to a lower level than usual. A study on dry saunas, published in the British Medical Journal, showed that: taking a sauna accelerated the uptake of insulin (a hormone that regulates glucose). This uptake, also favors skeletal muscle, which gives an explanation to the raise in hypertrophy with the use of saunas. A study on mice suggests that the reason muscle tissue is favored, instead of adipose tissue (or fat), is caused by an increase in the expression of a type of transporter known as GLUT-4 (responsible for transporting glucose from the blood into skeletal muscle). This is also possibly one of the reasons why heat acclimation has been shown to reduce the usage of muscle glycogen by about 40% to 50%, in addition to the increased flow of blood to the muscles.

Once again, saunas and heat exposure come with risks. Please read the disclaimer at the end of this article.

How to implement this:
—Most days when possible because saunas are awesome.
—In the morning after a night of non-optimal sleep
— After having had a large meal, waiting two hours or more to avoid going there with a full stomach.

2) Go for a walk, use a standing desk

As simple as the title suggests, especially after having a meal with my co-workers I would usually:

• Go for a walk.

• Use a standing desk.

I wish I had a treadmill desk so I could do both in the same time.

Note that this is not just about reducing glucose through the energy demands from standing or walking. It’s about keeping those levels stable. For example, in an office-based study published in the journal of Occupational and Environmental Medicine, working in a standing position seems to reduce spike in blood glucose by about 43% (compared to sitting for the same amount of time).

That said, some days I do sit for a long time either because I am focused on something, because I am on a meeting or just because I straight out do not want to. The key here is making sure those days are the exception not the rule.

How to implement this:
—Going for a walk after lunch at work, or after dinner.
—Using a standing desk during the day, when possible.

3) Eating bitter plants, herbs and spices

Of all the previous advice this one is probably the easiest. And the reason is because it tastes good. Before meals, including things such as: cinnamon, apple cider vinegar, lemon juice, or bitter melon can also be a way to keep stable levels of glucose. And if it is still early enough, green tea before or after meals can also increase insulin sensitivity and the uptake of glucose into the cells. That said, I have not done any controlled experiment to verify if these actually have an impact on how my glucose levels fluctuate.

But here’s the thing: there is enough evidence to support those compounds and they’re tasty anyway!

Ceylon Cinnamon, for example, is not only tasty but also seems to increase insulin sensitivity, and getting more glucose from the blood into cells. The other good news is, taking about 2 teaspoons a day would be enough you and do not need to ingest a whole cup or go through the cinnamon challenge to get those benefits. For example, In a randomized controlled trial, healthy individuals supplemented with either 5g of cinnamon or 5g of placebo. Consuming just 5g of cinnamon seems to reduce total plasma glucose responses to glucose with effects lasting up to twelve hours. A meta analysis shows that cinnamon also improves fasting blood glucose, total cholesterol, triglycerides and other biomarkers. So no need to wait for pumpkin spice season to use cinnamon.

Another plant extract that is worth mentioning is Berberine. When it comes to reducing blood glucose, a large number of studies suggests that the usage of berberine rivals oral hypoglycemic drugs such as Metformin in its effectiveness. Berberine seems to indirectly activate an enzyme known as AMPK while inhibiting PTP1B. As a result, this would increase the uptake of both carbs and fats into cells from the blood. However, I have not used Berberine myself because I am still skeptical about how it works on the molecular level and because most of the evidence comes from studies targeting individuals with type 2 diabetes. Please note that Berberine is known to inhibit the activity of certain enzymes which can lead to a host of drug interactions and that high doses can cause cramping and running pants.

I’ll personally stick with my Cinnamon on this one.

How to implement this:
—Adding bitters with food or before having a meal.
—Drinking green tea if it’s not late yet.

4 ) Using exercise throughout the day

By this I mean:

• exercising in a fasted state, before having any food.

• Taking small exercise breaks throughout the whole day instead of one hour of exercise (usually referred to as greasing the groove).

A study on the effects of fasted exercise (exercising before breakfast) suggests that exercising in a fasted state is better than exercising in a fed state in improving the tolerance to glucose and insulin sensitivity, in addition to facilitating muscle adaptations.

On the other hand, taking small exercise breaks during the day not only allows you to become proficient at certain movements (such as pull-ups, squats or dead-lifts), it also keeps your metabolism elevated throughout the day and frees you from the need of setting time for structured workouts.

Subscribe now

Important Disclaimer:
This post is for informational purposes only.
The content of this blog is not intended to be a substitute for professional medical advice, diagnosis, or treatment. You must consult with a health-care practitioner, before undertaking any of the exercises, habits, protocols, techniques or otherwise described herein.
This blog does not constitute the practice of medicine or any other professional health care services, including providing medical advice. The use of information on this post is at the user’s own risk. The author expressly disclaims responsibility for any adverse effects that may result from the use or application of the information contained herein.

Control blood glucose with these uncommon ways was originally published in Biohacker Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.

Time for another experiment. And this time it’s about sleep.

I have been using oral GABA to go faster to sleep. But I wasn’t sure it really helped. This is because the days I used GABA were days when I had trouble getting to sleep anyway, which made it hard to tell if GABA had any effect. So I decided to run an experiment and compare it to placebo.

After this experiment, I stopped using GABA as a supplement, and you will understand why in the results.

The form of GABA naturally present in the brain, Gamma-Aminobutyric Acid, is well established as an amino acid that supports in reducing stress, and enhancing sleep. But, using GABA as a supplement is not the same. Several human clinical trials have been conducted, results regarding the role of natural and/or biosynthetic oral GABA intake on stress and sleep seem to be mixed.

Summary of the results: GABA did not meet my expectations

I was expecting increased deep sleep, and total sleep, with GABA, but that did not seem to be the case. The results were:
➜ Less deep sleep compared to placebo, as measured by the Oura ring.
➜ Faster sleep compared to placebo, as measured by the Oura ring.
➜ Subjective feelings of drowsiness compared to placebo.

Note that the experiment is too small to truly make any conclusions on the statistical significance of the results.
That is why I stopped taking GABA for the purpose of enhancing sleep, but I still think there is potential for the use of GABA as a supplement for other purposes or in a different form than the existing supplements.

PS: if you’re interested in self-experimentation or want to get updates about the experiments I’m running feel free to join my mailing list here: https://ismail.bio/subscribe

Method: How I took the GABA supplement

How much GABA I took and how much time did it take to get a result.

• Which GABA supplement I took: Thorne’s PharmaGABA-100 supplement
  (note that this also contains: Calcium Laurate, Hypromellose Capsule, Magnesium Citrate and Microcrystalline Cellulose)

• How much GABA: 2 capsules with 100mg of GABA each (200mg in total).

• Placebo: I used acacia fiber as placebo. 2 x capsules on Placebo days.

• Blinding: I had a friend place the Placebo pills and GABA pills in similar jars, label them as A and B, and keep which one is A and which one is B under the jars (in a way that I could not see them). And for some extra “double blinding”, I did not check the meaning of A and B until I was done with the analysis. When this was done, I replaced the values in the visualisations below (GABA ended up being A, and Placebo being B).

• Timing: I took GABA 15min before going to bed.

Results

Before discussing the main results, it is good to note that there was no difference in bedtime and total sleep time in both experiments. The main results were that on days I used GABA, my deep sleep seems to be lower while it still seemed like I went faster to sleep.

Lower deep sleep with GABA supplementation

The first result that surprised me was that my deep sleep seemed lower with GABA intake. This goes contrary to a lot of the research that claims that GABA supplementation increases deep non-rem sleep. However as discussed in a systematic review [Hepsomali et al. 2020] the GABA supplementation studies fail to show any beneficial effect of GABA intake other than sleep latency (if at all).

The graphs here do not show statistical significance, but they give a better idea of how much difference there was between GABA and placebo.

Interpretation: why GABA may reduce deep sleep

I suspect this is due to the fact that after taking in GABA only increases plasma levels of GABA the first couple of minutes and then causes lower levels than the original baseline.

• ~30 minutes after supplementation: rapid increase of plasma GABA concentration. This may be responsible for the reduction in sleep latency.

• ~60 minutes after supplementation: a rapid decrease of plasma GABA concentration which may explain a lowering of deep sleep.

Faster to sleep with GABA

On the upside, I was getting faster to sleep with GABA. While as you can see there is an outlier value where it shows much longer, but even when this outlier is taken away (figure in the middle) it still seems that I slept faster with GABA.

This is consistent with my subjective feeling, with existing research and with the interpretation above.

Would I sacrifice deep sleep over faster sleep? It feels tempting, but most of the time I would prefer not to do that. Especially when there are other alternatives.

Possible confounding factors

Other things may have interfered with this self-experiment.

It is possible that my sleep on a weekend day is different from the rest of the weekdays. This experiment was only 12 days long, so we should check how the weekends and weekdays were split between GABA days and Placebo days.

As you can see, weekends consisted mostly of Placebo days, whereas weekdays consisted mostly of GABA days. It is possible that the results we saw earlier were impacted by that. In other words, that my deep sleep seemed longer during a Placebo day just because it was a weekend.

Without doing any complex statistics, the relationship between deep sleep and weekend/weekdays did not seem as high as the relationship between GABA and Placebo.

As I mentioned earlier, the following variables did not seem to differ : bedtime, wake-up time had no difference, neither did total time in bed. I also compared subjective measures of exercise intensity as these can also have an effect, but there was no major difference.

Conclusion

In this case both my gut feeling and the emotionless data agreed, and I stopped taking GABA.

This shows the benefit of experimenting yourself and not basing everything on existing research. What works for most people, does not work for a few people, and you may be part of the few people. Remember that this does not mean no one would benefit from GABA supplementation. If anything, there probably are a lot of people who would.

This experiment could be improved if it lasts about 28 days and if we design it to take into account all the possible confounding factors (exercise, weekends etc.).

If you have done any statistics in the past, you will know that this visual analysis is not enough to make any conclusions. But if you have done any self-experimentation, you would also understand that there is a sweet spot between what is statistically relevant and keeping your sanity.