Tertiary Structure Flashcards

Question

What are some characteristics of electron speed compared to X-rays?

Answer 1

* Most electrons orbit at a speed around 1/100th c (~2x106m/s). * In one X-ray cycle, an electron would travel 0.01Å, * X-rays see electrons as if they are standing still.

Answer 2

**Constructive interference:** * Two electrons oscillating in the same place * Oscillate in phase * Scatter twice as much. **Destructive interference:** * Two electrons separated by λ/2 * Oscillate out of phase * Cancel each other.

Answer 3

The "amplitude" of scattering is measured it is proportional to the differences of electron density in the direction of "reflection planes". The orientation and separation of reflection planes are determined by the directions of the incoming and scattered rays.

Answer 4

* To sample all angles. * About 100 X-ray diffraction images are required to provide a "data set". * Moving the X-rays and the detector gives a new set of planes. * Changing the angle of reflection changes the spacing (resolution).

Answer 5

**λ = 2d sinΘ** Explains the relationship between: * x-ray wavelength (λ) * angle of incidence (Θ) * spacing between planes (d) in crystallography.

Answer 6

To resolve features corresponding to small (atomic) distances in crystallography.

Answer 7

The x-ray waves can add up in phase (constructively) or out of phase (destructively), depending on the orientation of molecules with respect to the incoming and outgoing waves.

Answer 8

So that their diffraction patterns are in phase only at **discrete positions** that depend on the **internal dimensions** and **symmetry** of the crystals.

Answer 9

A 2D detector as discrete data points known as "reflections."

Answer 10

From x-rays "reflecting" off of discrete planes (Bragg planes) within the crystal, consisting of equivalent atoms of the structure aligned with the incident x-ray beam.

Answer 11

Crystal parameters and the wavelength of the incident x-rays, Relative intensities contain information about the structure of the molecule.

Answer 12

Improved techniques in: * crystal preparation (robotics) * synchrotron x-ray crystallography * software development

Answer 13

* Some proteins do not fold properly putside their native environment eg ones part of cell membrane * their structure is altered by interacting proteins or switcing between different states * This prevents crystal growth of gives structures that do not represent the natural structure of the protein

Answer 14

* Nuclear Magnetic Resonance (NMR) * Transmission electron microscopy (TEM).

Answer 15

Only technique capable of determining 3D structure of bio macromolecules at **atomic resolution in solution**

Answer 16

Best suited to large proteins or protein complexes

Answer 17

Used to predict the 3D structure of a new protein given the structure of at least one protein with the same fold. Technique is based on the strong tendency for structure and function to be conserved among homologous proteins.

Answer 18

1. Fold assignment 2. Target-template alignment 3. Model building 4. Model evaluation.

Answer 19

An AI-based protein structure prediction tool that predicts protein structures de novo (from scratch, without using previously solved proteins as templates).

Answer 20

Uses **Deep neural networks** that are trained to predict: 1. the distances between pairs of amino acids and 2. the angles between chemical bonds that connect those amino acids. The resulting scores are then optimized through a gradient descent resulting in highly accurate structures.

Answer 21

* Around **35%** of the more than 214 million predictions are deemed to be highly accurate, as good as experimentally determined structures. * Another **~45%** are considered to be accurate enough for many applications.

Answer 22

* Launched in 2021 * Contains AlphaFold-predicted models of protein structures of nearly the full UniProt proteome of humans and 20 model organisms, amounting to over 365,000 proteins.

Answer 23

* An AI-based protein structure prediction tool * Developed by researchers at Meta * Predicts the structures of ~600 million proteins * From metagenomic DNA from soil, seawater, and the human microbiome

Answer 24

* Uses a Transformer-based language model, ESM-2, which allows 3D protein structure prediction at the resolution of individual atoms. * Trained on the aa sequences of known proteins, fills in the gaps in missing sequence. * A second step combines this with information about the relationships between known protein structures and sequences, to generate predictions.

Answer 25

* Made up of ~617 million structures generated by ESMFold in just two weeks. * The vast majority are novel hypothetical proteins, coming from previously uncultured organisms – primarily bacteria and viruses. * Of the 617 million predictions, 225 million (approx. 1/3) are deemed to be high confidence predictions.