UNIT TWO

what are two problems that standard BLAST cannot solve?

• it cannot find homologs that are too distantly related

• it’s not built for (too detailed for) large queries (10,000 bp)

what does PSI BLAST stand for?

position specific iterated BLAST

what is the purpose of PSI BLAST?

used to detect weak but biologically meaningful relationships between proteins

What are the steps of PSI BLAST?

1. select a query and search it against a protein database

2. PSI-BLAST constructs a multiple sequence alignment then created a specialized position specific scoring matrix (PSSM)

3. the PSSM is used as a query against the database

4. PSI-BLAST estimates the statistical significance

5. repeat step 3 and 4 iteratively, typically five times. At each new search, a new profile is used as the query.

What does PSSM stand for?

position-specific scoring matrix

what are the steps to calculate the PSSM?

1. create a multiple sequence alignment

2. calculate raw frequencies - at each position the computer calculates how many times each of the 20 amino acids appear at each position

3. calculate overall frequency - divide the specific frequency by the overall frequency

4. log value of frequencies - find the log base 2 of all the values

what are and how to interpret the results of PSI-BLAST?

• iteration - the number of BLAST you are on

• number of hits

• number of hits > threshold - how many hits were significant

what does it mean to approach convergence?

as you move down the iterations, the rate of growth slows down, till no more new sequences are found

What is corruption?

presence of at least one false positive alignment with an e value < 10^-4 after 5 iterations

what are approaches to stopping corruption?

1. apply filtering of biased composition regions

2. adjust E value from 0.001 to a lower E value such as E = 0.0001

3. visually inspect the output from each iteration. remove suspicious hits by unchecking the box.

What is the result of one false positive?

one false positive can be amplified to many and it becomes permanent due to the nature of the math for the PSSM.

why proteomics?

-proteins are the functional (almost every diseases is the result of a protein failing not the gene)

-transcriptome information is only loosely related to protein levels

what does ELISA stand for?

enzyme linked immunosorbent assays

what is ELISA used to do?

used to detect and measure specific proteins or antibodies in a liquid sample.

steps of ELISA

1 - antibody is added; only binds to target protein

2 - secondary antibody is added; it looks for first antibody

3 - substrate is added

4 - enzyme catalyzes; produces color; the more color the more protein

steps of single protein analysis

1 - gel based separation: each black dot is a different protein

2 - spot excision: cut out one specific dot; same dot, same protein

3 - digestion: add on enzyme to chop up protein into smaller pieces called peptides

4 - MS analysis

what does protein sequence analysis allow for?

protein classifcation

What is the primary purpose of analyzing protein sequences using computational (in silico) methods?

characterize protein structures in silico and allows the prediction of protein structure and function

If a BLAST search shows high sequence homology between two proteins, what can you safely assume about their similarities?

may not have the same function but most always has the same structural fold

steps of shotgun analysis

1 - digestion of protein mixture

2 - liquid chromatography

3 - MS analysis

what are protein sequence databases?

-atlas of protein sequence and structure (currently known as Protein Information Resource (PIR)

-protein data bank

-UniProt

What can patterns of conservation in protein sequences tell us about specific amino acid residues?

They help determine which residues are under selective constraints, meaning those residues are critical for the protein's function

What is the defining characteristic of homologous proteins?

They share a common ancestor.

Different proteins evolve at ____ depending on their functional importance and structural requirements.

different

protein analysis is ____ sensitive than DNA analysis

more

How is the amino acid sequence of a protein typically generated for comparison?

It is generated from proteomics experiments

How is the % similarity between two protein sequences determined?

By aligning the two sequences and counting the number of identical residues or using an index of similarity (like a substitution matrix)

What is the primary method used to compare the 3D shapes of different proteins?

Superimposition (or structural overlay). This involves physically rotating and shifting the 3D models to see how well their "skeletons" match up.

pairwise alignment

multiple sequence alignment