Objective:

1. Learn basic concepts: amino acid structure, 3D protein visualization, and the variety of ML-based design tools.
2. Brainstorm as a group how to apply these tools to engineer a better bacteriophage (setting the stage for the final project).

Part A. Conceptual Questions

• Answer any of the following questions by Shuguang Zhang:

  • How many molecules of amino acids do you take with a piece of 500 grams of meat? (on average an amino acid is ~100 Daltons):

  Since 1 gr is around 6.022e+23 Daltons. Assuming a piece of meat has no fat and only protein will mean such piece of meat will have on average 3.011e+24 amino acids

  • Why humans eat beef but do not become a cow, eat fish but do not become fish?

  Humans do not become a cow or a fish because when proteins from food are digested, enzymes in the stomach and small intestine break them down into individual amino acids. These amino acids are universal building blocks shared by all living organisms. The human body then reassembles them into new proteins according to human genetic instructions, rather than those of the cow or fish

  • Why there are only 20 natural amino acids?

  The set of 20 standard amino acids is largely determined by the constraints of genetic encoding and evolutionary stability. The genetic code is optimized to efficiently translate triplet codons into a limited but diverse set of amino acids that provide chemical functionality while ensuring accurate and reliable protein synthesis. The available tRNA and ribosome machinery support these 20 amino acids in a way that balances diversity, stability, and efficiency in biological systems.

  • Can you make other non-natural amino acids? Design some new amino acids.

  .

  Yes, non-natural amino acids can be synthesized through chemical modifications or by engineering tRNA and ribosome systems to incorporate them into proteins. Some modifications include:

  1. Branched-chain amino acids with additional methyl or ethyl groups for altered hydrophobicity.
  2. Aromatic-ring modifications to enhance binding interactions in protein structures.
  3. Fluorinated amino acids for increased stability in harsh conditions.
  4. Metal-chelating amino acids that bind ions for enzymatic or structural applications.
  • Where did amino acids come from before enzymes that make them, and before life started?

  Amino acids likely formed naturally on early Earth from simple molecules present in the primordial environment. Experiments, such as the Miller-Urey experiment, have demonstrated that lightning, UV radiation, and other energy sources could drive the formation of amino acids from gases like methane, ammonia, and water. Additionally, some theories suggest that amino acids may have arrived on meteorites, as certain extraterrestrial samples contain amino acids.

  • If you make an alpha-helix using D-amino acids, what handedness (right or left) would you expect?`

  A left-handed helix would form because D-amino acids are the mirror image of the naturally occurring L-amino acids, which typically form right-handed alpha-helices.

  • Can you discover additional helices in proteins?

  Yes, in addition to alpha-helices, other helical structures such as pi-helices and 3_10 helices exist. These shorter or less common helices can be identified using advanced structural analysis techniques, including X-ray crystallography, cryo-EM, and AI-based structure prediction tools like AlphaFold.

  • Why most molecular helices are right-handed?

  Most naturally occurring amino acids are L-amino acids, which inherently favor right-handed helices. This handedness is also energetically favorable, as it optimizes hydrogen bonding and steric interactions, leading to more stable protein structures.

  • Why do beta-sheets tend to aggregate?
    • What is the driving force for b-sheet aggregation?

  Beta-sheets tend to aggregate because their structure allows for extensive intermolecular hydrogen bonding, leading to the formation of large, stable sheet-like assemblies. The driving force behind this aggregation is often the presence of hydrophobic amino acids, which promote hydrophobic interactions and further stabilize the aggregated state.

  • Why many amyloid diseases form b-sheet?
    • Can you use amyloid b-sheets as materials?

  Amyloid diseases involve beta-sheets because misfolded proteins can adopt an energetically stable beta-sheet conformation, leading to fibrillar aggregates that accumulate in tissues. These fibrils are highly ordered and resistant to degradation, contributing to diseases like Alzheimer’s and Parkinson’s.

  Yes, amyloid beta-sheets can be used as materials due to their high stability and self-assembling properties. They have been explored in tissue engineering, drug delivery, and biomaterials applications.

  • Design a b-sheet motif that forms a well-ordered structure.

Part B: Protein Analysis and Visualization

In this part of the homework, you will be using online resources and 3D visualization software to answer questions about proteins.

1. Pick any protein (from any organism) of your interest that has a 3D structure and answer the following questions.

   • Briefly describe the protein you selected and why you selected it.

   I picked Ferritin protein which play a crucial role in iron storage and metabolism. I chose it because it looks like a promising protein which could be used as a self-assembled liposome structure I could use for my final project of bio-sponge decorated with nano-body specific to the target which we are trying to remove. I specifically chose human Ferritin light chain for ease of design in my future application.

   • Identify the amino acid sequence of your protein.

     • How long is it? What is the most frequent amino acid? You can use this notebook to count most frequent amino acid - https://colab.research.google.com/drive/1vlAU_Y84lb04e4Nnaf1axU8nQA6_QBP1?usp=sharing

     The Ferritin light chain has 175 amino acids; Leucine (Leu, L) is the most frequent amino acid with 27 appearances.

     • How many protein sequence homologs are there for your protein?Hint: Use the pBLAST tool to search for homologs and ClustalOmega to align and visualize them. Tutorial Here
     • Does your protein belong to any protein family?

   • Identify the structure page of your protein in RCSB

     • When was the structure solved? Is it a good quality structure? Good quality structure is the one with good resolution. Smaller the better (Resolution: 2.70 Å)

     Structure was solved in 2006-07-11; it had resolution of 1.90 Å

     • Are there any other molecules in the solved structure apart from protein?

     There are two ligands mentioned: CADMIUM ION and SULFATE ION

     • Does your protein belong to any structure classification family?

     It belongs the the iron binding protein family

   • Open the structure of your protein in any 3D molecule visualization software:

     • PyMol Tutorial Here (hint: ChatGPT is good at PyMol commands)
     • Visualize the protein as "cartoon", "ribbon" and "ball and stick".

     cartoon

     

     ribbon

     

     ball

     

     sticks

     

     • Color the protein by secondary structure. Does it have more helices or sheets?

     The protein seem to have helices but no sheets

     

     • Color the protein by residue type. What can you tell about the distribution of hydrophobic vs hydrophilic residues?

     

     • Visualize the surface of the protein. Does it have any "holes" (aka binding pockets)?

     

     It looks like there pockets which can be used for iron (possibly)

Part C. Using ML-Based Protein Protein Tools

<aside> <img src="/icons/snippet_lightgray.svg" alt="/icons/snippet_lightgray.svg" width="40px" /> Resources: https://colab.research.google.com/drive/1hXStRY9VCyw52n17uWdWQBj__IcR2ztK?usp=sharing#scrollTo=38gFJBazNdzJ

</aside>

• Fold your protein with AlphaFold or ESMFold or Boltz and compare it to the real structure.
• Comment on:
  • Any predicted vs. experimental differences.
  • Low-confidence regions and why do you think they are low confidence?

Using alphafold server I was able to put it the sequence of 175 amino acids and go the following output:

it looks like the majority of the structure has high confidence (in blue) showing alpha helices matching with the experimental protein. The low confidence regions are likely regions of less stability where alpha helices which form a stable structure are higher in confidence.

ESM fold

• Inverse-fold your structure with ProteinMPNN

  • What sequence do you get?

  MSSQIRQNYSTDVEAAVNSLVNLYLQASYTYLSLGFYFDRDDVALEGVSHFFRELAEEKREGYERLLKMQNQRGGRALFQDIKKPAEDEWGKTPDAMKAAMALEKKLNQALLDLHALGSARTDPHLCDFLETHFLDEEVKLIKKMGDHLTNLHRLGGPEAGLGEYLFERLTLKHD

  • Is it the same as the original sequence you folded?
    • Why yes or no?

  When running pBLAST with the original sequence it is showing perfect macth- ProteinMPNN’s default behavior (T=0) already confirms that the original sequence is well-fitted to the structure. Higher temperatures allow for more divergence from the original sequence but the default behavior is good at predicting the original sequence.