Genomics: Gene Annotation Tutorial

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/19

flashcard set

Earn XP

Description and Tags

Comprehensive practice flashcards covering structural and functional gene annotation, ab initio models, algorithms, quality tiers, and critical exam traps based on the genomics tutorial.

20 Terms

1
New cards

What are the two primary layers of genome annotation, and what biological question does each layer answer?

Structural annotation answers "where" features are located (gene start and stop, exons, introns, UTRs, regulatory regions, repeats), while functional annotation answers "what" the features do (molecular function, biological process, cellular location, domains, disease links).

2
New cards

What three main evidence-based approaches are used for structural gene prediction?

Ab initio prediction (relies on intrinsic DNA signals only), homology-based prediction (uses known genes and proteins from databases like UniProt, RefSeq, or related species), and integrated prediction (combines ab initio, homology, and RNA-seq/EST evidence).

3
New cards
<p>In the gene annotation pipeline shown, which stage integrates ab initio signals, homology evidence, and RNA evidence before functional annotation?</p>

In the gene annotation pipeline shown, which stage integrates ab initio signals, homology evidence, and RNA evidence before functional annotation?

Integration (using tools such as MAKER and EVM).

4
New cards

Why does the Viterbi algorithm in ab initio gene finders typically fail to identify alternative splicing?

The Viterbi algorithm calculates and returns only the single most probable state path through the hidden Markov model (HMM), resulting in usually only one predicted isoform per gene locus.

5
New cards

What heuristic threshold defines an open reading frame (ORF), and why is this cutoff applied?

A threshold of ≥100\ge 100 codons (300 nt300\,\text{nt}) from an ATG start to an in-frame stop codon, applied because random non-coding sequences generate a stop codon roughly every 2020 codons (33 stops out of 6464 possible codons).

6
New cards

How do gene prediction accuracy rates compare between prokaryotes and eukaryotes, and what accounts for this difference?

Prokaryotes achieve >90%>90\% accuracy due to continuous, densely packed genes; eukaryotes achieve only 70–85%70\text{--}85\% accuracy due to introns, alternative splicing, and low gene density (1–2%1\text{--}2\% coding sequence in humans).

7
New cards

Which specific histone modifications from ChIP-seq data mark active promoters and active enhancers?

H3K4me3 marks active promoters, whereas H3K4me1 combined with H3K27ac marks active enhancers.

8
New cards

What are the standard rules of thumb for sequence identity percentages when inferring protein function via BLAST?

>70%>70\% sequence identity suggests similar function, 40–70%40\text{--}70\% indicates related function, and <40%<40\% leaves function uncertain.

9
New cards

What are the three aspects of Gene Ontology (GO), and how is TP53 mapped across them?

Molecular Function (e.g., GO:0003677 DNA binding), Biological Process (e.g., GO:0006915 apoptosis), and Cellular Component (e.g., GO:0005634 nucleus).

10
New cards

Why is Gene Ontology (GO) structured as a directed acyclic graph (DAG) rather than a strict hierarchy or tree?

Because a single GO term can possess multiple parent terms rather than being constrained to a single ancestral path.

11
New cards

What is the key functional distinction between PhyloP and PhastCons evolutionary conservation scoring tools?

PhyloP scores evolutionary constraint per individual nucleotide base and can identify accelerated evolution, whereas PhastCons identifies and scores multi-base conserved elements.

12
New cards

How do the three tiers of annotation quality (gold, silver, bronze) correspond to GENCODE and UniProt database levels?

Gold (manual curation with evidence) maps to GENCODE levels 1 and 2 and UniProt Swiss-Prot; silver (automatic annotation with evidence) and bronze (purely automatic prediction) map to GENCODE level 3 and UniProt TrEMBL.

13
New cards

What sequence of five evidence types makes up the "BDGSC" functional evidence ladder?

BLAST, Domains, GO, Structure, and Conservation ("Big Dogs Go Swimming Constantly").

14
New cards

What is the "golden rule" of genome annotation when functional characterization is uncertain?

"Hypothetical protein" is better than a wrong annotation.

15
New cards

How do assembly errors directly affect downstream structural gene annotation?

Annotation can never be better than the underlying assembly; gaps and misassemblies artificially split, truncate, or fuse gene models.

16
New cards

What is the mechanistic difference between signal sensors and content sensors in ab initio gene finders?

Signal sensors detect discrete consensus motifs (e.g., ATG start, stop codons, GT-AG splice junctions, TATA boxes), while content sensors evaluate statistical coding properties (e.g., codon usage, hexamer frequencies, 33-base periodicity, and GC content at codon position 3).

17
New cards

What is the major exam trap concerning data requirements for ab initio gene prediction?

Believing ab initio needs no data at all; although it requires no external experimental evidence for the target genome at prediction time, it strictly requires species-specific parameters trained on known gene models.

18
New cards

In which organisms or biological scenarios does the codon UGA code for an amino acid instead of signaling a stop?

In mitochondria and Mycoplasma, UGA encodes tryptophan (Trp); in selenoproteins, UGA is recoded to selenocysteine.

19
New cards

Why does an absence of RNA-seq reads at a genomic locus fail to disprove the presence of a functional gene?

Because RNA-seq expression is strictly conditional, depending heavily on the specific tissue type, developmental timing, or environmental stimuli sampled.

20
New cards

How is homology strictly differentiated from sequence similarity in genomics?

Homology is a qualitative, binary attribute indicating shared evolutionary ancestry (a gene is either homologous or not), whereas similarity is a quantitative metric representing percent identity or alignment score.