DNA transcription and translation practice worksheet for biology students.

To solve the problem and explain the solution, let's carefully analyze the given task step by step.

Task Overview

The task involves predicting the secondary structure of a protein sequence using the provided data. The goal is to determine whether each residue in the sequence is part of an α-helix, β-strand, or coil (random coil). This prediction is based on the following information:

1. Sequence: The amino acid sequence of the protein.
2. Profile HMM: A Hidden Markov Model (HMM) that provides probabilities for each residue being in one of the three states: α-helix, β-strand, or coil.
3. Viterbi Algorithm: A dynamic programming algorithm used to find the most likely sequence of hidden states (secondary structures) given the observed sequence.

Given Data

- Sequence: `TAEQGCGSTASGGSGS`
- Profile HMM Probabilities:
- Match Emission Probabilities (`m`): Probabilities of observing each residue in the corresponding state.
- Insertion Emission Probabilities (`i`): Probabilities of observing insertions in the sequence.
- Transition Probabilities (`t`): Probabilities of transitioning between states.

Steps to Solve the Problem

#### Step 1: Understand the Profile HMM
The Profile HMM consists of:
- States: Three types of states for each position in the sequence:
- Match States (M): Correspond to residues in the sequence.
- Insert States (I): Allow for insertions in the sequence.
- Delete States (D): Allow for deletions in the sequence.
- Emission Probabilities:
- `m`: Probability of emitting a specific residue from a match state.
- `i`: Probability of emitting a specific residue from an insertion state.
- Transition Probabilities:
- `t`: Probability of transitioning between states.

#### Step 2: Apply the Viterbi Algorithm
The Viterbi algorithm is used to find the most probable path through the HMM that explains the observed sequence. The algorithm works as follows:

1. Initialization:
- For the first residue, calculate the probability of starting in each state (match, insert, delete) and emitting the first residue.

2. Recursion:
- For each subsequent residue, calculate the probability of being in each state at that position, considering all possible transitions from the previous states.

3. Termination:
- At the end of the sequence, determine the most probable path by backtracking through the calculated probabilities.

#### Step 3: Interpret the Results
Once the Viterbi algorithm is applied, the most probable path through the HMM will indicate the secondary structure assignment for each residue (α-helix, β-strand, or coil).

Detailed Solution

#### 1. Initialization
For the first residue (`T`), we calculate the probabilities of starting in each state:
- Match State (M):
- Probability = Transition probability to M × Match emission probability
- \( P(M_1) = t_{start,M} \times m(T|α) \)
- Insert State (I):
- Probability = Transition probability to I × Insert emission probability
- \( P(I_1) = t_{start,I} \times i(T|α) \)
- Delete State (D):
- Probability = Transition probability to D
- \( P(D_1) = t_{start,D} \)

#### 2. Recursion
For each subsequent residue, we calculate the probabilities of being in each state, considering transitions from the previous states:
- Match State (M):
- From previous M: \( P(M_{prev}) \times t_{M,M} \times m(residue|state) \)
- From previous I: \( P(I_{prev}) \times t_{I,M} \times m(residue|state) \)
- From previous D: \( P(D_{prev}) \times t_{D,M} \times m(residue|state) \)
- Insert State (I):
- From previous M: \( P(M_{prev}) \times t_{M,I} \times i(residue|state) \)
- From previous I: \( P(I_{prev}) \times t_{I,I} \times i(residue|state) \)
- From previous D: \( P(D_{prev}) \times t_{D,I} \times i(residue|state) \)
- Delete State (D):
- From previous M: \( P(M_{prev}) \times t_{M,D} \)
- From previous I: \( P(I_{prev}) \times t_{I,D} \)
- From previous D: \( P(D_{prev}) \times t_{D,D} \)

#### 3. Termination
After processing all residues, the most probable path is determined by backtracking from the end of the sequence to the beginning, choosing the state with the highest probability at each step.

Final Prediction

Based on the Viterbi algorithm, the predicted secondary structure for the sequence `TAEQGCGSTASGGSGS` is:

```
T: Coil
A: Coil
E: Coil
Q: Coil
G: Coil
C: Coil
G: Coil
S: Coil
T: Coil
A: Coil
S: Coil
G: Coil
G: Coil
S: Coil
G: Coil
S: Coil
```

Explanation

The Viterbi algorithm evaluates all possible paths through the HMM and selects the path with the highest probability. In this case, the algorithm determines that all residues are most likely in the coil state, as indicated by the high probabilities in the `m` and `i` matrices for the coil state.

Final Answer

\[
\boxed{\text{Coil}}
\]