Giulia Magri Ribeiro

Ensaio opcional(10/03)

The use of the central dogma of molecular biology isn't so useful to understand/create historical reconstructions. The central dogma of molecular biology is an explanation of the flow of genetic information within a biological system. This dogma has also been stated as DNA, makes RNA and RNA makes protein. A historical reconstruction is a “tree” that infers evolutionary relationships between species. Historical reconstructions are built based on morphological or genetic characteristics. But we need to be careful of how we understand these genetic characteristics. When we think in a individual, the central dogma can explain the most common ways of flowing information in its life. But evolutionary events are deeper than that. For understanding the flow of genetic information in historical reconstructions we need to consider events like, for example: epigenetics, prions or lateral transference. Epigenetics can interfere in morphological characteristics of organisms and even in their fitness, all this without changing the base sequence that you use to build the trees. Lateral transferences is a kind of genetic information flow that can make the work of building trees harder if you don't identify this event. Lateral transferences can occur in different organisms that are not necessarily related, and you are seeing just an homoplastic character. For these reasons, phylogenetic studies in general are using combinations of genes and morphological characters, trying to obtain the phylogenetic signal for the groups. So, if you ignore all these different cases of flowing of genetic information, and stays with the central dogma of molecular biology you are basically staying in individual level, and not working with all the possibilities of evolutionary events that can be happening in your group.

Achei o texto claro e direto. Com parágrafos curtos e lógicos. Você falou sobre como mecanismos diferentes devem influenciar na reconstrução histórica. Faltou falar sobre as divergências do próprio código genético em si.

Ensaio 1 (17/03)

Alignment is an art. You could identify many evolutionary processes and mistakes just looking to your data. The methods that we have to build historical reconstructions are based on dichotomies that came from duplication models of DNA replication. However, this dichotomies doesn’t consider reticulation events, for example. Reticulation events, like lateral transferences, can disturb the “vertical” signal (phylogenetically related). The tip here is, look to your alignment before you use a pipeline. Pipelines will just analyze your data and give you the tree. It is interesting to understand what your pipeline gives to you before trust it.

Comentário (Marcos)

Olá Giulia.

Não tenho certeza se a sua primeira frase é um “topic sentense”, apesar de ter um único verbo na frase. Sua frase não vai direto ao ponto como sugerido pelo Daniel. Não ficou clara a relação entre “art” e os processos (mecanismos) de alinhamento.
Essa é minha única observação. Suas frases são curtas e me parecem muito boas. Elas, também, atendem as orientações do Daniel.


Ensaio 2 (24/03)

Genes used as phylogenetic markers must be neutral. Phylogenetic analysis uses substitution rate to infer relationships. We can underestimate or overestimate substitutions rates. These mistakes can occur when your site is not neutral. Sites under selective pressures are usually underestimated. That happens, because you have purifying selection acting on it. Molecular markers usually are neutral. However, in big data times it is difficult to know if everything is neutral or not. Because of that, or you understand the genes you are using or you find a way to escape from this bias. New methods can discard these “not neutral” data. Otherwise, a way to escape could be using aleatory sites to reconstructions.

Correção: Victor Calvanese
Texto muito bom, acredito que você tenha captado bem as ideias propostas em aula para construção lógica do assunto. Posso sugerir, entretanto, que explore mais as informações e conclua o ensaio de forma mais contundente.

Ensaio 3

Público alvo: alunos estudando deriva e seleção.

What is responsible to create complex adaptations in nature? One of the classic examples of natural selection has as main character Darwin’s finches. Their diversity lies mainly on size and shape of their beaks (Abzhanov, 2010). Results seems to show that environmental pressures causes this variation (Grant, 1976). Usually the process pointed as responsible for this variation is natural selection. After the modern synthesis, we have started to make the link between complex adaptations and their genetics ( When, this question if natural selection is the only cause emerge. If a dry cause the death of finches. Their effective population size diminished. Other process may have contributed to the establishment of a difference in variation (grant,1976). One thing to test is the role of genetic drift in that situation of low population size.


Ensaio Opcional (27/04)

OBS: Achei que o meu ensaio anterior havia sumido então refiz ele. Porém ele reapareceu. :(

Darwin finches are a traditional example of natural selection. But new mechanisms of data collection could bring new insights to this question. One thing that many researchers test was the influence of random effects in this evolutionary process. The finches history started with a dry season. All ecosystem was affected by this long dry. Finches reproduction was affected. They were usually reproducing after the rain. Many plants had died. The small, soft seeds were all gone. Most part of the finches died. This could be characterized as a strong bottleneck in the population. And this, was what bring this question about random effects to many researchers. Many tests were made. Some of them brought the importance of drift in many closely related Galápagos finches species. Today the consensus are that both selection and drift had a strong contribution to many aspects in this process.


Comentários (Rosana Cunha)

"Darwin finches are a traditional example of natural selection, but new mechanisms of data collection could bring new insights to this question. One thing that many researchers have tested is the influence of random effects in this evolutionary process. The finches history started with a dry season. All ecosystem was affected by this long dry. Finches reproduction was affected. They were usually reproducing after the rain. Many plants had died. The small and soft seeds were all gone. Most part of the finches died. This could be characterized as a strong bottleneck in the population. And this, was what bring this question about random effects to many researchers. Many tests were made. Some of them brought the importance of drift in many closely related Galápagos finches species. Today the consensus are that both, selection and drift, have a strong contribution to many aspects in this process."

- Texto bem objetivo e direto, sentenças curtas, de fácil compreensão, boa conclusão.
- A "topic sentence" teria ficado mais completa e instigadora se você tivesse colocado a vírgula ao invés do ponto.
- Apesar das sentenças curtas serem incentivadas para melhor entendimento do leitor, seu texto ficou um pouco telegráfico, ok?
- Utilizar mais conectivos para dar fluidez à leitura.

Ensaio 4

Público alvo: Professores ensino básico

Phylogenetic trees are appearing in increasing numbers in biology textbooks (2). Trees are helpful to structure knowledge of biological diversity. However, the tree way of thinking is not easy to get. It needs to be constantly exercitated (3). Other difficulty is that we usually find misconceptions and anthropocentric trees in textbooks. This does not do more than confound the students. However, if it is so difficult to explain and understand trees even for the teachers why this tool is so useful? In 1960, initiates in science the period of phylogenetic systematics (1). Phylogenetic systematics introduce the tree thinking tradition. It is through this way of thinking that we can understand classification. In this way of thinking, organisms are just part of a single tree (2). Evolution, in its core, is a claim that species are related by descent in a common ancestry. What I am proposing is that, the easiest way to a student to understand, for example, vertebrates evolution, should be through phylogenetic trees and plotting evolutive novelties. When the student does that, he can abandon the essentialist vision and he can understand that these novelties could be base to new groups.


1. O'hara R. J. 1993. Systematic generalization, historical fate and the species problem. Systematic Biology, 42:231-246.
2. Baum D. A. & Offner S. 2008. Phylogenics & tree-thinking. The American Biology Teacher, 70:222–229.
3. Sandvik H. 2008. Tree thinking cannot taken for granted: challenges for teaching phylogenetics. Theory in Biosciences, 127:45–51.

Comentários por Lucas D. de Campos

Seu parágrafo está sucinto e bem claro quanto a ideia do auxílio das árvores filogenéticas no entendimento da evolução.
Achei a pergunta da frase “ However, if it is so difficult…” um pouco confusa. Talvez por ser um pouco extensa. Talvez, se você escolher entre “explain” ou “understand” imagino que ficaria um sentença mais fluida. Nada demais.
Texto bem legal!

Ensaio 5 (12/05)

Público: Usuários de árvores filogenéticas

There are many reasons to build a tree (1). Any comparison between sequences assumes evolutionary relationships. Phylogenetic trees describes evolutionary relationships between sequences (1). Evolutionary trees showed us the three domains (2). Evolutionary trees showed that cases of encephalitis in New York and New England represented the first examples of the mosquito-borne West Nile virus in the western hemisphere (1). Evolutionary trees can solve questions since profound relationships between groups until just identification organisms or sequences. Nowadays we have many methods in our hands to use. Why we can’t for example, use the NJ tree as a fast starting point for a phylogeny. Each kind of analysis have your explaining power. Each kind of question should have your optimality criterion. The only exigence here is to know until where the method you choose explains your question.

definition of oprmality criterion:
1. HOLDER, Mark; LEWIS, Paul O. Phylogeny estimation: traditional and Bayesian approaches. Nature reviews genetics, v. 4, n. 4, p. 275-284, 2003
2. WOESE, Carl R.; KANDLER, Otto; WHEELIS, Mark L. Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya. Proceedings of the National Academy of Sciences, v. 87, n. 12, p. 4576-4579, 1990.

Comentários por Bruno (19/05): Olá Giulia. Achei o objetivo do seu texto claro, isto é, apresentar um resumo de porque e como fazemos árvores filogenéticas. Porém, achei que ele ficou meio telegráfico, isto é, as frases ficaram curtas (como o Daniel já tinha nos alertado, devemos evitar construções muito longas) porém muito distintas; isto é, não há uma ligação entre elas. Isso se percebe na quantidade de vezes com que você teve de repetir a expressão "árvores filogenéticas". Há também alguns probleminhas gramaticais e algumas palavras usadas fora de contexto. Mas no geral, gostei do seu

Ensaio 6

Schools are part of systematics. Each one of these schools have your philosophies and assumptions. However, the paradigm of in which one of these schools we believe changed with time. It was common to use typological methods, when we had nothing but our eyes to compare things. At this time, we classified organisms by their similarities. UPGMA or phenetics is a method to classify by general similarities. However, if the objective of systematics is to characterize and understand the formation process of the diversity of life. Instead of compare similarity, we need to compare the evolution of the character. This is what phylogenetic schools do.

Ensaio 7

Público: Pessoas que estão pensando em modelos.

In science usually simple is better (Elliot sober, Aeon, 2016). We can connect this idea to the logic concept of the Ockham’s razor. Ockham’s razor is a philosophy principle to solve problems. If you have two explanations for a phenomenon, the one that have fewer steps to produce the result is better. We should not admit more causes of natural things than the sufficient to explain the phenomena (Elliot sober, Aeon, 2016). We can make a connection of these ideas in the model selection. We should not admit more parameters than the sufficient to explain the history of our group. Each new parameter we assume in our model brings a new biological assumption to our history. If you bring 12 different probabilities for nucleotide substitutions, probably you will enhance a little your maximum likelihood. However, what you are either saying is that every kind of substitution have a biologically different probability to happen. For that purpose, you really need a theoretical base. That is why selection model strategies (AIC), choose the model that can explain your data equally with fewer number of parameters.

"Why is simpler better?." Aeon, 3 May 2016

"AIC Scores as Evidence: A Bayesian Interpretation." In M. Forster and P. Bandyopadhyay (eds.), Philosophy of Statistics (Handbook of the Philosophy of Science, Volume 7), 2011, Oxford: Elsevier, pp. 535-549.

Comentários de Victor Calvanese: texto muito bem escrito, seguindo as orientações propostas pelo professor. Embora bastante coerente, em minha opinião, você poderia ter explorado mais sua conclusão.

Ensaio 8 (02/06)

Público alvo: Pessoas interessadas em modelos de construção de árvores

Tree searching algorithms can be stuck in local optima (Giribet, 2007). There is a large number of possible trees (Giribet, 2007). That configures the tree searching space. Maximum likelihood is a way to estimate the most plausible tree of this space. Tree spaces are infinite. It is possible to find out suboptimal trees and equally optimal trees. The problem starts because: traditionally the exploration of the tree space is through hill climbing algorithms. Algorithms like that stops as soon as it reach a local maximum likelihood optima. These algorithms could not go back to see if there is another place equally good or better. They can´t because to go to that place it need to pass to worse scored places. However, there is a possible solution to this. That solution could be tell the algorithm after find the local optima, go backwards and try to go to a little different place and see if this improve the score of the final tree.

Comentários (Rosana Cunha)

- Boa "topic sentence", seu texto flui a partir desta ideia principal.
- Sua linguagem ficou muito telegráfica Giulia, apesar das ideias serem boas, não há elementos conectivos entre as frases e a leitura acaba não fluindo bem.
- " Maximum likelihood is a way to estimate the most plausible tree of this space", você pegou isso do artigo? Tem que referenciar esta informação.
- Acho que faltou uma pequena conclusão

Ensaio 9 (9/06)

Molecular clocks could be strict or relaxed. The molecular clock is a technique that uses the mutation rate of molecules to deduce the divergence time between lineages. The existence of the molecular clock phenomena occurs first than the explanation to it. Pauling observed this phenomenon with protein sizes between lineages that diverge in different timing and after with the development of the neutral theory we had the explanation to this phenomenon. Neutral theory explains saying that the major part of the substitutions are neutral and occurs at a constant rate. That rate is used to calibrate clocks. However, constant rates are not mandatories in nature. Many genes experience higher or lower levels of purifying selection. Many genes, have diverged long time ago and are already saturated. Saturation occurs when enough time have passed and sites have more than one change that you couldn’t detect. Molecular clock users developed a relaxed clock solution to that kind of problem. Those models take into account that kind of variation across lineages to estimate the divergence times.

Ensaio 10 (23/06)

O formato da disciplina é muito bom. Eu já conhecia esse formato anteriormente, pois já havia feito a disciplina da graduação de principais tendências evolutivas de microrganismos eucariontes. Gosto de matérias que priorizem discussões à aulas teóricas. Gosto muito do exercício de escrita ao final da aula. Isto por que, apesar de eu ter dificuldade na escrita, acho importante exercitá-la. Acho que a única crítica que eu tenho é em relação aos temas da primeira avaliação. Alguns destes temas eram muito gerais ou tinham muita sobreposição entre si e era difícil saber qual era a diferença entre alguns dos temas.
Acredito que me esforcei pela disciplina. Por eu gostar do formato, tentei o máximo possível ler os textos. As vezes, como era colocado no dia anterior eu não conseguia ler todos. Mas tentei escrever meus textos em inglês. Li o livrinho de técnicas de escrita para tentar melhorar. Apesar de eu acreditar que eu ainda tenha muita dificuldade. Tentei reler minha prova 1 e pensar no que eu faria para melhorá-la. Então acredito que a medida do possível tive um bom desempenho na disciplina.
Talvez para os próximos anos seria legal aprofundar mais algumas aulas, principalmente da primeira parte de genética de populações. Como não são temas que envolvem técnicas, talvez seja possível exercitar mais a aplicação desses conceitos no estudo de evolução molecular. Eu sempre tentei refletir sobre isso nos meus ensaios, mas acho que teriam sido boas discussões para a aula.
Gostei de ter aprendido um pouco mais sobre a lógica da construção de árvores filogenéticas. Que apesar de não ser algo que eu use muito. É importante para entender o trabalho de outras pessoas inclusive.
Acho muito válido esse exercício dos alunos corrigirem os ensaios dos colegas. Porém senti que acabou virando algo protocolar. Uma obrigação. Então, uma sugestão é fazer as pessoas se sentarem juntas e discutirem os ensaios uma da outra.

Auto avaliação: Pela minha dificuldade na escrita e por falar pouco em aula 0,95

Giulia Ribeiro

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License