A functional selection model explains evolutionary robustness despite plasticity in regulatory networks

Its out on the Molecular Systems Biology website. Yay!

This paper summarizes quite a bit of work by Naomi and Ilan. It started with "here is a short 2-week project" and ended as big part of Naomi's PHD and a massive paper.

Highlights (from our synopsis):

By tracing the evolutionary history of transcriptional networks across 23 fungi, two seemingly contradictory trends are observed: rapid target turnover and conserved function. This is reconciled by a model that invokes strong selection to conserve the overall function of a motif, but not its individual targets.

• The vast majority of cis-regulatory elements in genes’ promoters are rapidly gained and lost across species.
• Despite this rapid turnover, most transcription factors are associated with a conserved function even at great evolutionary distances.
• A functional selection turnover model reconciles these two phenomena by invoking a preference to conserve the overall function of the motif but not the individual target genes.
• Our model fits the variation in measured transcription factor binding profiles across species in both yeasts and mammals.

Systematic Dissection of Roles for Chromatin Regulators in a Yeast Stress Response

The monster is out! 

A work that started more than 2 years ago, finally resulted in a publication. The paper came out in PLoS Biology. This work is tight collaboration between Ollie's lab and ours, with Assaf and Hsuiyi as first co-authors. Avital and Ayelet also played an important part and are coauthors on the paper.

This publication is also the first time we publish experimental work done in our own lab. We can say that we are officially an experimental lab! Figure 2 (below) is all about microscopy measurements that we collected on our system.

Behind the cover

Getting a cover image for a journal was a first for me. As such I thought it would be interesting to revisit the behind the scenes. I am going for a short recap.

Initially when the paper was accepted the editor encouraged us to sugget a cover image. We (actually the Broad people, Manfred, Brian, and Aviv) recruited Sigrid Knemeyer who collected ideas and came up with some concept images. We (the three mentioned above, Moran, and me) then had several rounds of emails with Sigrid where each of us made comments, said what we liked or not about the options. Sigrid came up with a more detailed concept, which we sent to the editor. The editor wasn't too excited about it, and so we went back to the drawing board.

This time Sigrid, came up with four initial concepts.

We really liked some elements of each, but thought that the vortex of letters does not capture the idea of short reads. The stacks and stacks of papers was also nice, but didn't make it clear that the reads are short.

The next concept was to replace stacks of papers with short strips, like fortunes (as in fortune cookies), and combine it with the vortex.

We really liked this, but the final touch was adding the assembled sequence coming out of the vortex. This and change of color tone resulted in the concept that we sent to the journal.

They really liked it, and sent it to their graphic editors who came up with the final version. 

As you can see, they dispensed with the fortune-cookie strips, but kept the letters in "words" that capture the idea of short fragments and made other visual changes.

In conclusion, this was an interesting process. Sigrid was great in making this happen, and deserve a lot of credit. Thanks!

We made the cover!

Our paper on transcript assembly appeared online few weeks ago. Now the actual issue came out, and we made the cover, with a striking image.

The idea came out of iterations between all of us (Manfred Grabher, Brian Haas, Moran, Aviv and me) and  Sigrid Knemeyer who is an artist/illustrator working at the Broad Institute. Sigrid was great in developing the concept with our input of rough concepts, and this whirlwind was the final outcome.

Substantial Histone Reduction Modulates Genome-wide Nucleosomal Occupancy and Global Transcriptional Output

The recent wave of publications seem to continue, and another paper just appeared in PLoS Biology. Like the previous one, this one was in the reviewing rounds for very long, and so we are very happy that it finally appeared.

This project is a collaboration with the group of Alessandra Agresti and Marco Bianchi from San Raffaele University in Milan. They are part of a European consortium that we are also members of, and they approached us to get help on dealing with nucleosome organization. Assaf got involved, into what was dubbed "The Italian Job" in lab meeting. After almost two years, the result is an impressive article.

To summarize the basic idea, the experiments by Barbara Celona, the main author, and other member's of the Milan group show that deletion in a key gene reduces the number of nucleosomes in the genome. As a result the DNA strands are packed by smaller number of nucleosomes. By mapping the locations of these nucleosomes we can see if nucleosome reduction leads to uniform change in density, or preferential reduction at particular locations, and what is the effect on transcription.

Source: PLoS Biology Synopsis by Robin Mejia

Our results show that nucleosome loss is mostly in locations that are less packed in normal cells, and moreover, that this loss correlates with increased transcription.

Cool!

Patterns and Mechanisms of Ancestral Histone Protein Inheritance in Budding Yeast

Hot from the (virtual) press! A paper that we have been working on for almost a year appeared today in PLoS Biology. This is three way collaboration between Fred van Leeuwen's group at the Netherlands Cancer Institute, Ollie Rando's group at UMass Medical School and our group. The work on our end was mostly Assaf's who is a co-first author. 

The interesting aspect of this paper is that using a trick developed in Fred's group, they can switch a "tag" on a histone H3 protein. This means that up to a certain point all H3 proteins are tagged by one tag, and after the switch they are all tagged with a different tag. Using this switch we can find out where these "old" histones are several generations after the switch. This gives us insight on how the cells perserve old histones during several cycles of replication.

To quote ourselves:

To our surprise, ancestral histones accumulate near the 5′ end of long, relatively inactive genes. Using a mathematical model, we show that our results can be explained by the combined effects of histone replacement, histone movement along genes from 3′ towards 5′ ends, and histone spreading during replication. Our results show that old histones do move but stay relatively close to their original location (within around 400 base-pairs), which places important constraints on how chromatin could potentially carry epigenetic information. Our findings also suggest that accumulation of the ancestral histones that are inherited can influence histone modification patterns.

To get to these conclusions Assaf implemented a mathematical model that takes into account the processes that affect nucleosome positions, and then showed that this model can provide a good fit to the data if we consider turnover, passback of nucleosomes (from 3' to 5'), and localized dispersion during replication.  

Full-length transcriptome assembly from RNA-Seq data without a reference genome

Just hot off the press - a new paper appeared in advanced online publication in Nature Biotechnology. This paper describe a new computational pipeline to recover RNA transcript sequence and abundance (aka the "transcriptome") from RNA-seq data without using the genome as a reference.

This paper is a joint work of Moran from our group and Brian Haas and Manfred Grabher, both from the Broad Institute. The three of them (Brian, Manfred and Moran) developed a three-part tool to handle massive number of sequencing reads and assembling them to accurate account of the transcriptome. They also applied a very thorough evaluation of our method and how it compares to most of the state of the art methods in the field. This evaluation  by itself is of interest to the this emerging area of analysis.

The basic problem is that RNA-seq returns many (millions) of sequences of different fragements of the original RNA molecules from the sample. To make sense of it, we need to assumble it (like puzzle) into longer pieces. The two general strategies for doing so are nicely illustrated in this figure (from a review by Haas and Zodie):

The "straightforward" approach is to align reads to the reference genome, and then use this mapping to guide reconstruction. The less obvious approach, that we took here, is to first assemble the puzzle, and then map to the genome. This turns out to be often as accurate (or even more), since it is less suceptibles to problems in mapping to the genome, differences between the reference genome and the actual sample, and partial/fragmented references.

Our strategy is based on three steps, each processing the data very efficiently while maintaining information and dealing with sequence errors and rare events.

Metabolic labeling of RNA uncovers principles of RNA production and degradation dynamics in mammalian cells

An "online" version of a paper appeared in Nature Biotechnology. This work of Michal Rabani, a Hebrew University graduate who is now a PhD student in Aviv Regev's lab at MIT. Michal is also a member of our group, spending several months a year here. This work also included Ido Amit, a postdoc at Aviv's lab who is starting a new lab at the Weizmann later this year.

In this work Michal and Ido used a new method to label newly synthesized RNA using a modified version of uridine that can be later used to separate it from the total RNA pool. This way they measured and sequenced "new" RNA. Michal then used these measurements to estimate the production rate of RNA in cells that respond to an external stimuli. Moreover, by contrasting the production rate of RNA and total RNA she could reconstruct what are the degradation rates through the process.

The main takeaway message from the paper is that RNA degradation rate (or stability) differs between different RNA species. However, for most genes, it did not change dramatically during the response. Thus, most of the shape of RNA level was regulated by production. We did find that for some genes changes in degradation rate was important to modulate the response.

Comparative Functional Genomics of the Fission Yeasts

Fresh off the press: An online version appeared on the Science Express website (pre-publication online release of Science articles). This is a genome paper where the main result is the sequencing of a genome. In this case, actually two new species, both related to the fission yeast, S. Pombe. 

(figure from Nick Rhind's website)

The fission yeast is well studied model organsim. As its name suggests, it divides in half during growth (unlike the budding yeast). It is also closer to humans than the budding yeast (aka Baker's yeast) in terms of many mechanisms. And so, many important discoveries in fission yeast in the areas of cell cycle, silenced genomic regions (heterochromatin) and other regions, were more directly related to mammalian cells.

(figure from http://physiology.uvm.edu/lord )

The new world of genomes allows to compare related genomes and use these comparisons to understand what part of the genome are functional. Such comparative genomics has proved immensely fruitful. However, so far very few related species of S. pombe were sequenced. The sequencing of the new genomes, and functional annotations of the genes in the these two genomes provide fresh insights into S. pombe genetics. 

As you can see in this evolutionary tree that contains most of the currently sequenced yeast genomes, the "neighborhood" of S. cervisiae (bakers' yeast) and C. albicans (a species of Candida, a human pathogen) are well populated (horizontal branch lengths correspond to evolutionary distance). With the addition of S. japanicus and S. octosporus, the branch leading to S. pombe is now better explored.

(figure courtesy of Naomi and Ilan)

This project was spearheaded by Nick Rhind (UMass Medical School) and Chad Nusbaum (Broad Institute). Our contribution was in helping make sense of functional data and comparative analysis. Moran helped in identifying transcribed genes in the two new species based on samples collected at Aviv Regev's lab. She discovered many antisense transcripts that regulate some of the key genes in developmental decision making in these yeasts. Naomi and Ilan used comparative methods to find regulatory elements in these genomes, and showed how regulatory elements that are conserved throughout the tree of ascomycota yeasts change their function in these yeasts to match their life style.

Exploring transcription regulation through cell-to-cell variability

Yay! A paper that was in the works for two years, including more than six months in review, finally appeared in Proceedings of National Academy of Sciences, USA.

This describes Ruty's MSC dissertation, which was done with help from Ariel and me. In this work she took a very large dataset collected by Martin Jonikas and Maya Schuldiner at Jonathan Weissman's lab, and reanalyzed it to uncover new findings.

The key point of the paper is the use of "noise" or variability in expression levels of the same protein between genetically identical cells as a phenotype for genetic screen. Ruty shows that mutations that lead to abnormally high or abnormally low variability are involved in key processes in regulating transcription. Moreover, by using double knockouts she can pinpoint some mechanisms on these defects.

This work is cool as it shows that promise for the general theme that we want to pursue in the lab, of using genetic perturbation and examining effect on protein expression in single cells.

Regulation of Blood Cell Differentiation

Another cause for celebration. A project that was active for several years has made it to publication. Today a paper with Noa as a first author appeared in Cell. This project was mostly done when Noa was at Aviv Regev's lab at the Broad Institute. She analyzed the expression profiles of genes through the differentiation of hematopoitic cell lineages --- how a bone marrow stem cell develops in to different types of blood cells.

                                                                          (Source: Cell)

The interesting part is that this analysis uncovered multiple "modules" --- groups of genes --- that are reused in different lineages. This shows that differentiation is much more complex than previously assumed about this system.

Congratulations to Noa!

You are what your father ate?!?

I have been lax at reporting papers coming out, but decided we need to change that. So here is a new type of post - talking about new publication.

Today, Cell published a paper "Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals" by Carone BR*, Fauquier L*, Habib N*, Shea JM*, Hart CE, Li R, Bock C, Li C, Gu H, Zamore PD, Meissner A, Weng Z, Hofmann HA, Friedman N, Rando OJ. 

This is culmination of a wild roller coaster ride over the last few years and months. The project has been one of Ollie's (Rando, OJ, that is) for quite few years and we were brought on to help.

For some reason Cell put a press release with the title "You are what your father ate" which seems a bit of an over statement. But it definitely managed to get the press interested. So far there is press coverage by the Time, and various other interesting venues, such as Pravda.

The basic story, is that Ollie and later his students have been feeding mice either control diet or one that is low on protein and high on suger. These mice where then mated with females and then removed from their cage. Ollie then examined gene expression in the offspring livers and found that they were different depending on the father. In particular, lipid metabolism in these mice was different.  

(source: Cell)

Our involvement was in trying to help Ollie make sense of the data. As usual he found the cool stuff by himself but needed reaffirmation from computational people :-) Naomi (Habib, N, that is) analyzed the expression data to make sure that the results made sense. She also compared the list of genes that changed in the offsprings to many relevant datasets in the literature and found connections that helped understand the nature of these changes. She also analyzed other datasets, including small RNAs in these livers, and gene expression in sperms from fathers of the two different groups. 

All by all, this project was one of the main things Naomi did over the last few years. It is great to have it come out, and in particular in such a prestigious venue.