Week 1 HW: Principles and Practices

1. Project Proposal

I’m fascinated by slime molds because they are rather unique - being amoeba-like single celled microorganisms that can fuse into one macro-scale organism to find food and spread spores more efficiently and demonstrate some level of intelligence.

There’re quite many studies and research around slime mold and the organism is known in popular culture, however, many people still think that slime mold is a fungus that can somehow solve mazes :)

I want to develop a slime mold open source incubator and experimentation platform for growing, studying, and sharing data and experiment protocols; to make it approachable to grow and play with slime molds for anyone who is interested by just picking up samples from the nearest park or forest, and have scientific potential for citizen-science, similar to iNaturalist.

The platform would consist of an autonomous closed box with computer (like Raspberry Pi), camera, and sensors, elements for controlling temperature and humidity, and possibly masked light/UV to create arbitrary light shapes, a feeding pipette, and electrodes.

The software would control environment, camera, lights, feeder, etc. via API and web interface. With current state of computer vision and reasoning models, it is also approachable to implement an AI-aided loop that will recognize slime mold parameters, record them, and even autonomously optimize environment for better growth to help novice users not kill their slime mold pet.

Users will be able to receive notifications, view time lapses and photos, design and execute experiments via web interface and/or API. Experiment results and protocols could be uploaded and shared on a web hub. Users would be able to check other people’s results and download their protocols to try.

I think such project could be a nice sci-pop device for anyone interested in biology and at the same time generate data that can be useful for scientists. E.g., with enough users of the platform, a science lab can ask people to execute their protocol to gather statistics, and this will be a win-win situation.

2. Governance / policy goals

The main governance goals of this project are to ensure that the platform contributes to safe and constructive use of biology experimentation and does not create harm for users or the environment.

Goal 1: Popularizing biology and DIY experimentation Lower barriers for people to safely experiment with biological systems, make biology more accessible and interesting, and encourage responsible hobbyist and educational use of living organisms.

Goal 2: Creating a standard and safe environment for biological experiments Provide a safer alternative to random home experiments with petri dishes and unknown organisms by creating a controlled, closed and reusable environment for experimentation with biological material.

Goal 3: Building a shared experiment knowledge base and helping science Enable sharing experiment protocols and results, create a community around experimentation, and potentially help scientists collect data via citizen-science and distributed experiments.

3. Potential governance actions

Option 1: Make safety the top priority in platform design

Purpose: What is done now and what changes am I proposing?

Right now many biology hobby experiments happen using open containers or improvised setups, where safety depends mostly on user behavior. The proposed change is to make safety the main principle of the platform: a closed device, safe disposal and cleaning procedures, and software safeguards become part of the default design and policy of the open source project.

Design: What is needed to make it work?

Safety principles would become part of the project “constitution” and design guidelines for hardware and software. Device should minimize direct contact with biological material, support safe cleaning and disposal, and include mechanisms to detect contamination and warn users or even have a self-cleanup mode, e.g., with UV light. Open source contributors and builders would be encouraged to follow these design principles.

Assumptions: What could I have wrong?

Users and contributors might ignore safety recommendations. Technical solutions for contamination detection or safe automation might be unreliable or too complex to make.

Risks of Failure & “Success”

If safety design becomes too complex or expensive, fewer people may build or use the platform. If successful, users might develop a false sense of security and start unsafe experiments anyway.

Option 2: Engage teachers, scientists and hobbyists as early adopters

Purpose: What is done now and what changes am I proposing?

Right now you can find different protocols and experiments online, but typically there’s no guaranteed way to reproduce the experiment reliably or get feedback from experiment creators. The suggested platform is better to start with people who already have experience with safe biological experiments and education and can create initial content so that others can follow and reproduce on their devices. Web hub will allow sharing results and asking questions.

Design: What is needed to make it work?

Encourage teachers, bio clubs, scientists and experienced hobbyists to try the platform first, publish safe experiment protocols, and establish good usage practices. These early adopters define baseline safe experimentation and act as examples for later users.

Assumptions: What could I have wrong?

Assumes educators and scientists are interested and have time to participate, and that new users will follow practices defined by early adopters.

Risks of Failure & “Success”

Platform may fail to attract enough early adopters. If successful, community norms might become too rigid and discourage experimentation.

Option 3: Allow experiments and protocols sharing and build self-regulating community

Purpose: What is done now and what changes am I proposing?

Right now experiment results with slime molds might be shared within bio clubs or on hobbyist websites or not recorded at all. The core idea of the platform is to allow experiments and results to be shared so that users learn from each other and build a living community around safe experimentation. Additionally, the database of experimental data can be used by real scientists.

Design: What is needed to make it work?

Users can upload experiment protocols and results, try experiments created by others, and discuss or report unsafe practices. Community feedback and reputation mechanisms help highlight good protocols and discourage unsafe ones.

Assumptions: What could I have wrong?

Assumes enough active users exist for community moderation to work and that users behave responsibly.

Risks of Failure & “Success”

Platform might not reach enough scale to self-regulate. On the other hand, users might blindly copy experiments without understanding risks.

4. Governance options scoring

5. Prioritization and trade-offs

I think for the early stage of the project, the main actions are Option 1 and Option 3, while Option 2 plays an important role in later stages of the platform (that might never occur :)).

Safety-by-design (Option 1) is necessary to reduce risks from the beginning and make the platform acceptable for schools and hobbyists. Sharing experiments and building community (Option 3) is essential for long-term usefulness and citizen-science potential. Option 2 is important to establish safe practices early, but in order to get there, the project needs to be well established already.

The main concern is balancing safety and complexity: making the device safer may increase cost. Another concern is that openness might create risk of misuse, since open sharing makes experimentation easier but also allows unsafe ideas to spread.

The big uncertainty is whether it will be possible to stabilize slime mold in such an incubator and reliably/reproducibly control its behavior and if it’s possible to build certain hardware features cheaply.

Week 2 Lecture Prep

Homework Questions from Professor Jacobson:

1. Nature’s machinery for copying DNA is called polymerase. What is the error rate of polymerase? How does this compare to the length of the human genome. How does biology deal with that discrepancy?

The error rate is about one mistake for every million bp copied, according to slide 8. The length of human genome is ~3 billon bp, so it should be making ~3000 mistakes. However, there’re error correction mechanisms like proofreading and mismatch repair that reduce error rate to ~1:10^9-1010. Errors are still made.

2. How many different ways are there to code (DNA nucleotide code) for an average human protein? In practice what are some of the reasons that all of these different codes don’t work to code for the protein of interest?

Average protein 1036bp ~ 345 amino acids, which given that amino acid can be coded in 3 different ways on average, makes lots of DNA combinations. Only small fraction of them can be actually built. Some of the reasons for this are: a) cells might lack rare codons, so the process stucks, b) protein folding itself blocks the process.

Homework Questions from Dr. LeProust:

1. What’s the most commonly used method for oligo synthesis currently?

Phosphoramidite solid-phase synthesis.

2. Why is it difficult to make oligos longer than 200nt via direct synthesis?

Errors rate grows with the length.

3. Why can’t you make a 2000bp gene via direct oligo synthesis?

Also because of errors accumulation. This is why it is easier to make short blocks and glue them together.

Homework Question from George Church:

Choose ONE of the following three questions to answer; and please cite AI prompts or paper citations used, if any. [Using Google & Prof. Church’s slide #4] What are the 10 essential amino acids in all animals and how does this affect your view of the “Lysine Contingency”?

Essential are: Phenylalanine, Valine, Threonine, Tryptophan, Isoleucine, Methionine, Histidine, Arginine, Leucine, Lysine.

With regards to “Lysine Contingency”, it turns out that: a) dinosaurs (like birds) should likely already had such lysine contingency without genetic modification. b) they could get lysine from “normal” food like other animals, so it’s not such a contingency after all.