ppgardner-lecture03-genomesize-complexity.pdf
0 likes75 views
The document discusses genome size, complexity, and the concept of 'junk DNA,' highlighting the c-value paradox that genome size does not correlate with organism complexity. It references key lectures and the ENCODE project, which aims to map functional elements of the human genome. The document also outlines controversies surrounding the functional definitions of non-coding DNA and presents various sources of information and critiques related to these topics.
ppgardner-lecture03-genomesize-complexity.pdf
- 1. GENE315: Genome size and complexity Paul Gardner March 14, 2023
- 2. Introduction ▶ I started at Otago in 2018. ▶ Taught bioinformatics, genomics & biostatistics for 6 years at the University of Canterbury. ▶ Worked for ∼ 10 years in Europe. I’ve lived in Bielefeld, Copenhagen and Cambridge. Taught courses in Denmark, Cambridge, Poland, Portugal, Australia, USA, and NZ. ▶ Originally from Whāngārā, married with three children.
- 3. Overview of lectures 03-08 ▶ 3. Genome size and complexity. ▶ 4. Comparative genomics & genome annotation. ▶ 5. Comparative genomics & sequence alignment. ▶ 6. Comparative genomics & homology search. ▶ 7. Measuring genome function.
- 4. Objectives for lecture 03 ▶ An appreciation for: ▶ “Junk DNA” & the C-value paradox. ▶ The ENCODE project. ▶ Critical appraisals of genome function.
- 5. What is a gene? NIH: Wikipedia:
- 6. Ohno S (1972) So Much “Junk DNA” in our Genome. Brookhaven Symposium on Biology. – available on Blackboard.
- 7. Summary of Susumu Ohno’s main points about junk DNA ▶ The human genome is roughly 3 × 109 bases, ≈ 750 times E. coli. ▶ Assuming the same gene density as E. coli, humans will have ≈ 3 million genes. ▶ HOWEVER, if gene density is preserved, then salamanders have 10 times the number of genes that we do. ▶ There may be a strict upper limit to the number of gene loci. ▶ Therefore, only a fraction of human DNA is “genic”. ▶ Argues that based upon mutation rates, that the number of genes must be less than 1 mutation rate to avoid meltdown. ▶ and the number of genes is ≈ 30, 000, therefore about 6% of our DNA is genic.1 ▶ Speculates about some hypothetical functions of non-coding DNA, e.g. chromosome structure, promoter/operators, inhibiting non-sense/frameshift mutations, ... ▶ Speculates that our genome is littered with the “fossil” remains of extinct genes.
- 8. Vertebrate genomes & junk DNA ▶ Vary in length by two orders of magnitude, e.g. bird genomes are ≈ 1Gb, while salamander genomes are ≈ 32Gb (giant lungfish genome is ≈ 43Gb). ▶ Variation largely driven by decaying remnants of transposons. ▶ ≈ 300Mb of sequence is conserved across the vertebrates. ▶ Randomly generated sequences when inserted into genomes are also transcribed, promoters are very degenerate. ▶ I.e. many different sequences can be bound by a transcription factor ▶ Which suggests the number of functional elements should not scale with genome length. Ohno S (1972) So much’junk’DNA in our genome. In Evolution of Genetic Systems, Brookhaven Symp. Biol.
- 9. Mammalian genome variation & junk DNA ▶ But salamanders are special... (if the genome sizes were longer in the birds than in the salamanders we could rationalise how complex birds are). ▶ Let’s look at just the mammalian extremes of genome size then... the red vizcacha rat, the mountain vizcacha rat and a bat! Which of these mammals is more complex? Southern bent-wing bat Miniopterus orianae bassanii ~1.6 Gb genome Mountain viscacha rat Octomys mimax ~3.2 Gb genome Plains viscacha rat Tympanoctomys barrerae ~6.2 Gb genome Evans et al. (2017) Evolution of the Largest Mammalian Genome. GBE.
- 10. 30 years later we forgot about Ohno’s line of reasoning... Editorial (2000) The nature of the number. Nature genetics. Willyard (2018) New human gene tally reignites debate. Nature. Hatje et al. (2019) The Protein-Coding Human Genome: Annotating High-Hanging Fruits. BioEssays.
- 11. Genome sizes & number of genes 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 A Nucleus Cytoplasm ER Golgi Lysosome Mitochondria B Interactions Genome length ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● Paramecium Arabidopsis Homo Gemmata Escherichia Saccharomyces Lokiarchaeum Mycoplasma Nanoarchaeum Escherichia phage 104 105 106 107 108 109 1010 101 102 103 104 105 ● Animal Fungi Plant Protists Archaea Bacteria Phage Transect Number of protein coding genes ▶ Prediction: “In fact, there seems to be a strict upper limit for the number of genes loci which we can afford to keep in our genome” (Ohno, 1972) ▶ Eukaryotic genomes can vary widely in size, with little change in the number of protein-coding genes. Gardner, Gumy & Fineran (2019) Unpublished.
- 12. Related to junk DNA, is the “C-value paradox” ▶ C-values (a.k.a. genome size) vary enormously between closely related species. ▶ Vertebrates: birds (1 Gb), bats (1.6 Gb), human (3 Gb), tuatara (5 Gb), salamander (32 Gb), lungfish (130 Gb). ▶ Allium (onions): 7 Mb to 32 Mb. ▶ There is little relationship between the “perceived complexity” (or the corresponding number of genes) of a species and the size of a genome. ▶ Given that C-values are relatively constant within species/cells, yet bears no relationship to the complexity, or presumed number of genes, then this somewhat of a paradox. https://en.wikipedia.org/wiki/C-value#C-value paradox Eddy, SR (2012) The C-value paradox, junk DNA and ENCODE. Current Biology. Palazzo & Gregory et al. (2014) The Case for Junk DNA PLoS Genetics.
- 13. Drivers of genome size variation C-value variation is largely driven by transposable elements and other repetitive DNA elements https://sandwalk.blogspot.com/2018/03/whats-in-your-genome-pie-chart.html http://www.dfam.org/entry/DF0000001/hits
- 14. Protein functions, 2023 – still lots of unknowns cytoskeletal protein 3% (627) transporter 5% (1031) scaffold/adaptor protein 4% (746) protein−binding activity modulator 4% (793) RNA metabolism protein 4% (827) gene−specific transcriptional regulator 7% (1516) defense/immunity protein 3% (664) metabolite interconversion enzyme 9% (1939) protein modifying enzyme 8% (1622) transmembrane signal receptor 6% (1151) Unclassified 33% (6695) Other 14% (2980) Unknown 0% (0) 20,851 Human Proteome Functions 2023 Gardner (2023) Data from Panther: http://www.pantherdb.org
- 15. The ENCODE Project 2012 ▶ 443 authors, from genome/analysis centres around the world ▶ Aim to map functional elements across the human genome ▶ Generated LOTS of important data that we are still using: ▶ Transcription: RNA-seq, CAGE (Cap Analysis Gene Expression), RNA-PET (paired end tag) ▶ Translation: mass spectrometry ▶ Transcription-factor-binding sites: ChIP-seq and DNase-seq ▶ Chromatin structure: DNase-seq, FAIRE-seq, histone ChIP-seq and MNase-seq ▶ DNA methylation sites: RRBS assay ▶ Largely used 3 cell lines: erythroleukaemia K562; B-lymphoblastoid GM12878 & H1 embryonic stem cells The ENCODE Project Consortium (2012) An integrated encyclopedia of DNA elements in the human genome. Nature.
- 17. ENCODE 2012: LOTS of cool results (30 papers) E.g. transcription is correlated (r = 0.9) with transcription factor binding (left) and cancer associated SNPs are linked to chromatin structure, which may influence gene expression (right).
- 18. ENCODE 2012 ▶ Acknowledged that “Comparative genomic studies suggest that 3% to 8% of bases are under purifying (negative) selection” ▶ Yet conclude “The vast majority (80.4%) of the human genome participates in at least one biochemical RNA- and/or chromatin-associated event in at least one cell type.” ▶ “Many non-coding variants in individual genome sequences lie in ENCODE-annotated functional regions; this number is at least as large as those that lie in protein-coding genes.” ▶ “Single nucleotide polymorphisms (SNPs) associated with disease by GWAS are enriched within non-coding functional elements”
- 19. The ENCODE press lead to some “interesting” headlines:
- 20. And there was a predictable scientific backlash: ▶ Eddy, SR (2012) The C-value paradox, junk DNA and ENCODE. Current Biology. ▶ Q&A style discussion of C-values, junk DNA and ENCODE ▶ Eddy, SR (2013) The ENCODE project: Missteps overshadowing a success. Current Biology. ▶ “all reproducible biochemical events were claimed to be ‘critical’ and ‘needed’.” ▶ “Far from disproving junk DNA, ENCODE’s operationalized definition of function included junk DNA”. ▶ “genomes contain a lot of DNA that is not critically important for host functions, much of which arises from mobile element replication”. ▶ Sean proposes the “The Random Genome Project: the missing negative control”. ▶ “Biology is noisy” therefore an assessment of noise in transcription, translation & DNA binding would allow an assessment of how much of the ENCODE results are noise, and how much is genuinely functional. ▶ Recent progress: Deciphering eukaryotic gene-regulatory logic with 100 million random promoters
- 21. And further responses ▶ Doolittle WF (2013) Is junk DNA bunk? A critique of ENCODE. PNAS. ▶ If the number of functional elements were to rise significantly with C-value then ▶ (i) organisms with larger genomes are more complex phenotypically ▶ NB. Organism/phenotypic complexity is generally not well defined. ▶ (ii) ENCODE’s definition of a functional element identifies many sites that would not be considered functional or phenotype-determining by standard uses in biology ▶ (iii) the same phenotypic functions are often determined in a more diffuse fashion in larger-genomed organisms ▶ Graur D et al. (2013) On the immortality of television sets:“function” in the human genome according to the evolution-free gospel of ENCODE. Genome biology and evolution. ▶ “a very forcefully worded critique” – W. Ford Doolittle ▶ “angry, dogmatic, scattershot, sometimes inaccurate” – Sean Eddy
- 22. Note: Transcription & translation are likely to be noisy processes Genome Functional Junk Untranscribed Transcribed Transcribed Untranscribed Untranslated Translated Translated Untranslated Transcriptome Proteome
- 23. Promoters/Transcription factor binding motifs are degenerate ▶ “Degenerate” motifs refers to the fact that many different sequences can promote transcription ▶ This implies there is lots of potential for random, noisy transcription http://jaspar.genereg.net/
- 24. How would you determine if a genomic region is “functional”? ▶ Lecture #7
- 25. Humans are not the most (or least) complex species! Big brains, hands, sure. But we can’t breath underwater water, regrow limbs, don’t have tentacles, can’t fly unassisted, can’t photosynthesise, melt under mild radiation, turn inside out in a vacuum, get cancers - circulatory disease - infectious disease, do stupid things, get too hot and too cold, ... Humans are weak and puny! Great Chain of Being https://en.wikipedia.org/wiki/Category:Obsolete biological theories
- 26. The main points ▶ C-values (genome sizes) vary dramatically, even between closely related species. ▶ The “C-value paradox” is that genome size is not proportional to the perceived organism “complexity”. ▶ Many “functional” non-coding elements exist in the genome (.e.g ncRNAs, promoters, etc.) ▶ The majority of the eukaryotic genomes is littered with “Junk” DNA. ▶ these are largely pseudogenised transposable elements and endogenous retroviruses ▶ NB. junk ̸= in-active AND non-coding ̸= junk ▶ ENCODE is a valuable resource for identifying “biochemical” activity in cell lines. ▶ Some of the ENCODE conclusions and press are controversial. ▶ Reports of the death of junk DNA are greatly exaggerated.
- 27. Self-evaluation exercises ▶ Outline the C-value paradox, include in your answer a definition of junk DNA. ▶ What are the main sources of junk DNA? ▶ Describe the ENCODE project and main conclusions. ▶ Outline an approach to distinguish between functional and noisy cellular processes. ▶ The below figures have been used to associate proportions of noncoding DNA with “complexity”. Using your knowledge of the C-value paradox, write a critique of this result. Mattick (2004) The hidden genetic program of complex organisms. Scientific American. Taft et al. (2007) The relationship between non-protein-coding DNA and eukaryotic complexity. BioEssays.
- 28. Self-evaluation exercises ▶ Use chatGPT to generate an essay on “genome function and junk DNA”. Critique the essay based upon the references presented in this lecture. AI generated art by https://creator.nightcafe.studio/studio, “Junk DNA”, Preset: Surreal
- 29. Suggested reading ▶ Important: Ohno S (1972) So Much “Junk DNA” in our Genome. Brookhaven Symposium on Biology. ▶ Important: Eddy, SR (2013) The ENCODE project: Missteps overshadowing a success. Current Biology. ▶ Extra: Palazzo & Gregory (2014) The Case for Junk DNA. PLOS Genetics.
- 30. Questions relating to my lectures can be asked & viewed here: https://docs.google.com/document/d/1PQd dp7C 0cXA8SwUv- qrkTOj8c8fUAt-U Z5dg2yc8/edit?usp=sharing
- 32. The End Karitane, 23 Oct 2020.