Lecture 2

Question 1

When comparing two nucleotide sequences what do we need to keep in mind?

Answer 1

We have to keep in mind that they are result of mutation during replication (genotypic level) and selection (phenotypic level)

Question 2

Could you give and example of how evolution occurs in genotypic and phenotypic level ?

Answer 2

At sequence level is GAC changes to AAC, this muting results in antibody not binding anymore and there fore it minds to HIV instead.

Question 3

How many nucleotides are there for RNA and DNA?

Answer 3

DNA: TCAG, RNA: UCAG

Question 4

How many amino acids are there?

Answer 4

20

Question 5

What is a codon?

Answer 5

3 nucelotides which encode for one amino acid

Question 6

To change the phenotype, at least how many nucleotides should be changed?

Answer 6

1

Question 7

What could have to DNA/RNA when they are copied?

Answer 7

mutation, insertion, deletion, repeat, inversions, inverted repeated

Question 8

To compare two sequences, what do we need to know about them initially?

Answer 8

which positions in the sequences correspond to each other.

Question 9

A correct alignment represents — events such as — and — ( — and —)

Answer 9

actual, substitutions, indels, insertion and deletion

Question 10

Sequences with shared ancestry are referred to as what?

Answer 10

homologous

Question 11

to be able to align sequences we base the idea of what?

Answer 11

That there is a common ancestor which genes evolved from. The ancestor has a certain nucleotide or aminoaide at a certain position which could have changed during the evolutionary history.

Question 12

Could we be sure of the alignment?

Answer 12

no but we choose the model with the highest probability or score

Question 13

What are types of alignments?

Answer 13

pairwise : protein-protein, DNA/DNA, RNA/RNA, DNA or RNA with a protein : shifts within a codon
multiple sequence alignment

Question 14

What are two strategies for alignments? explain each

Answer 14

Global: aligns one sequence to there other from start to end, local: finds the longest subsequences with highest similarity

Question 15

what are strategies for finding a pairwise sequence alignment?

e.g different methods

Answer 15

qualitative method : dot-matrix method
exact method via dynamic programming: needle-man for global and smith for local,
heuristic and fast methods: word methods like BLAST

Question 16

align CTG and CTAAG, CTAAGAAG and CTAAG, ATC and CTAAG using the dot matrix method and say what each pattern shows

Answer 16

slide 15

Question 17

What are the pros and cons of dot matrix?

Answer 17

pro: visually easy to identify sequence features such as indels, repeated, inversions and inverted repeats, cons: time consuming and due to being qualitative doesnt give optimal alignment

Question 18

In quantitative methods how do we know to accept for instance mutation or gap?

Answer 18

By assigning costs to different actions in the alignment process

Question 19

What are 3 possibilities of characters being compared at one position?

Answer 19

match, mismatch, gap

Question 20

What are the total number of alignments between a sequence of length m and sequence of length n<m? explain how you got to this

Answer 20

slide 21

Question 21

What is dynamic programming?

Answer 21

its breaking down a more complicated problem to simpler sub problems and solve them in a recursive manner, and finding the optimal solution to the sub problems.

Question 22

Explain how the smith waterman algorithm works? how many steps does it include?

Answer 22

if we have two sequences A and B, we calculate the optimal alignment to one point only once and we built am trip for the two sequences with seaA row and serb column. the field (I,j) corresponds to the score of the optimal alignment with the nucleotides ai and bj as the end of the alignment. Finally we find the best way through the complete matrix
it requires m*n steps

Question 23

What are the pros and cons of the smith waterman algorithm?

Answer 23

its fast in comparison of brute force , it finds the optimal or one of multiple alignments with the same highest score, its only for local alignments, only pair wise alignment possible, still too slow for scanning against big libraries

Question 24

align AATC and AGAC according to needle man wunsh

Answer 24

slide 31c

Question 25

How does BLAST work?

Answer 25

it aligns the query sequence against all known sequences from genback which is a collection of DNA sequences and it reports the best alignment

Question 26

Explain in detail how the BLAST algorithm works.

Answer 26

1. split the query into sequences of length k 2. search k-letter words in the database sequences, similar is ok but is scored. 3. only keep the sequences above the set threshold 3. expand the k-letter words to the right and left. 4. expand the k-letter words to the right and left 5. stop if the score drops below a certain threshold 6. keep only the pairwise alignments which are above the threshold 7. report the database sequences (doesnt have to be exact sequences could be similar)

Question 27

What are the pros and cons of blast?

Answer 27

up to 50-100 times faster than direct alignment, allows searches for exact matches but also similarity up to a pre-defined threshold. cons: doesnt guarantee the optimal pairwise alignments of the query and database sequences, you can only find a match when the gene/sequence is available in the database.

Question 28

Why do we need MSA?

Answer 28

to reconstruct the evolutionary history of individuals in a phylogeny, to assess the sequence conservation of proteins

Question 29

What do the columns in the MSA represent?

Answer 29

It represents amino acids or nucleotides which have descended from the same position in the sequence as in the common ancestor

Question 30

What do we want in an MSA?

Answer 30

A MSA which correctly represents the evolutionary history of a set of sequences

Question 31

Name and explain two approaches for MSA, including their advantage/ disadvantages.

Answer 31

1- Ad hoc : define a reference strain for the genome and pairwise align all sequences with the reference strain. advantage : position numbering is the same for each sequence disadvantage : only possible when one knows which species the sequences come from 2- extending smith waterman to more dimensions: Instead of having a 2D matrix, we will need a k -dimensional matrix , additionally we need a scoring scheme which accounts for all sequences at each alignment position. disadvantage: is extremely slow for instance k sequences with length m require m^k steps 3- different algorithms such as Muscle, Malign which are implemented in alignment viewers such as aliview

Question 32

What are the weakness and strenghts of dotmatrix method, smith waterman , needleman-wunsch and BLAST?

Answer 32

Dot matrix : gives visual hints for indels but doesnt give optimal alignment smith and needle: faster and give optimal blast: not optimal

Question 33

With Which alignment methods do you get an optimal alignment?

Answer 33

Needleman and smith