- 1 The Hottest Topics in Science Today: a Graduate Student's Perspective
- 1.1 Identifiers for the 21st century: How to design, provision, and reuse persistent identifiers to maximize utility and impact of life science data 
- 1.2 TZAP: A telomere-associated protein involved in telomere length control 
- 1.3 A heterochromatin-dependent transcription machinery drives piRNA expression 
- 1.4 Disruptions of Topological Chromatin Domains cause pathogenic rewiring of gene-enhancer interactions 
- 1.5 Cohesin Loss Eliminates All Loop Domains 
- 1.6 HSP90 Shapes the Consequences of Human Genetic Variation 
- 1.7 Nucleic Acid Detection with CRISPR-Cas13a/c2c2 
- 1.8 Adorable Cat Photos
The Hottest Topics in Science Today: a Graduate Student's Perspective
Join me in reading some of the trendiest scientific discoveries of 2017! I will endeavor to provide some stimulating questions every Tuesday to expand your own thinking about the article. Other helpful resources are available in the Materials Section. Check out even more perspectives by going to the Discussion Section.
Identifiers for the 21st century: How to design, provision, and reuse persistent identifiers to maximize utility and impact of life science data 
- As it stands, scientific writing still includes cumbersome written citations. The authors have correctly indicated that many scientific journals are moving towards unique identifiers, like DOIs, that allow citations to be better tracked. What might this mean for the future of scientific citations in publications? What might this mean for our future "online presence" as scientists?
- The authors focus on avoiding embedding meaning in identifiers. This is something that we, as scientists, tend to rely on when coding our data for ease of retrieval. These are lessons that are often highlighted for computer science majors and those that are learning coding: should these classes be a part of future curriculum in PhD programs to better prepare future scientists?
- Large datasets are becoming more common as big hypothesis generating experimental techniques, like RNAseq, are more available. Currently these datasets are not easily accessible or uniformly stored across groups. The authors provide several suggestions for making big data more readily sharable. How could the scientific community better communicate standard within fields where "big data" is becoming more common? Who is responsible for setting these standards?
13:17, 10 October 2017 (PDT)nice Qs for today's expert faculty to address.
TZAP: A telomere-associated protein involved in telomere length control 
- Given what we have learned last week, what might the impact of the authors renaming Kruppel-like zinc finger protein ZBTB48 as telomeric zinc finger–associated protein (TZAP)? Was this justified? How might this impact the field moving forward? What about other scientific fields?
- The authors conclude that TZAP directly binds DNA via its zinc finger domains (Figure 2). Does their data support direct binding through these domains? If so, how do they demonstrate it? If not, how might they demonstrate it?
- Much of what we know about DNA replication is provided by in vitro assays. How could the authors examine TZAP function in vivo?
A heterochromatin-dependent transcription machinery drives piRNA expression 
- Why are some piRNA clusters Rhino dependent while others are Rhino independent?
- In figure 5, the authors seek to bypass the need for Moonshiner to recruit TRF2 to Rhino by using an inventive construct; however, this construct does not fully rescue piRNA expression. What might account for the partial rescue? How might the authors have interpreted this data differently? What other methods might they have used to address this question more fully?
- The authors note that Rhino is not conserved outside of drosophilids, but that TRF2 is conserved. Given what they have uncovered about the complicated nature of TRF2 recruitment in their work, what implications might their work have for other insects or mammals? What are some future directions for those studying piRNA clusters in other species, like mice?
Disruptions of Topological Chromatin Domains cause pathogenic rewiring of gene-enhancer interactions 
- The mice that the authors created to mimic the F1 family inversion died shortly after birth and did not have any obvious phenotypes that recapitulated the human disease that corresponded to this genetic defect; however, the authors continued to use these animals in their analyses. Does this seem appropriate? Why or why not? Why might these mice not have developed the phenotype that we see in humans despite the other dysregulation that the authors report later in their paper (eg Wnt6 expression in the distal mesoderm)?
- These TADs are highly conserved and seem extremely important for regulation. Do you think it is likely that gross dysregulation of TADs would often be embryonic lethal?
- This paper changes the way we might think about how mutations impact gene regulation. These deletions, duplications, and insertions alter chromatin structure. What other types of modifications impact chromatin structure that could account for some of these phenotypes?
Maureen 15:18, 6 November 2017 (PST) yes that's what I wonder about (Q3).
Cohesin Loss Eliminates All Loop Domains 
- This is less of a discussion question and more of a point that is important to note regarding accurate reporting in scientific works. The authors have titled this work "Cohesin Loss Eliminates 'All' Loop Domains", but proceed to spend a significant portion of the paper discussing and examining cohesin-independent loops(they are even represented in the graphical abstract of the article.) What is true the purpose of the title of a journal article? Are titles held to the same rigorous standards of precise reporting as the rest of the body of work? This title could be taken to be a clear overstatement of the authors findings, a misrepresentation of the research reported therein, or even an entirely inaccurate depiction of the role of cohesion based upon the authors' own evidence. Is there pressure for authors to produce "click-bait" like titles for articles? How does might the climate of funding and publish or perish contribute to this?
- In Figure 1C and 1D, the authors assert that loss of cohesion does not affect CTCF binding; however, CTCF binding does appear to be reduced (some peaks are lost on chr1 and there does seem to be reduced binding depicted.) The authors do not address this, probably because this was not significant nor the main focus of the paper. What might account for the reduction of some of the CTCF binding in the context of the authors' findings regarding the role of cohesion on loop domain structure?
- Interestingly, the authors find that cohesin-dependent loop domains interfere with links between the superenhancer elements described by the authors. Why might this be biologically important? What role might this aspect of loop formation play in regulation in the cell? How else might cohesion-dependent loop formation be used as a regulatory agent in the cell?
- Why did the authors decide to use PRO-seq rather than using GRO-seq? PRO-seq offers resolution on an individual base scale, but why did the authors require this information? How might they have used this information or what might they use this information for in the future? What does this tell you about how research is conducted?
HSP90 Shapes the Consequences of Human Genetic Variation 
- Are there other chaperone proteins that would be worth examining beyond HSP70 or HSP90?
- The authors briefly allude to other disorders that might involve chaperone buffering. How could their findings be used to explain some of the heterogeneity seen in genetic disorders? What about the complexity of compounding mutations? Could there be mutations in HSP70 or HSP90 that affect their ability to "buffer" in these disorders?
- The authors findings have implications about testing some variants of unknown significance. How could we exploit the authors' findings to create a test that might be useful for screening?
Nucleic Acid Detection with CRISPR-Cas13a/c2c2 
- I hadn't heard about RPA before this publication. The authors use this as a part of SHERLOCK, but I'm wondering why we are not using RPA in place of qPCR? I would like to learn more about this method specifically.
- The authors harp on the usefulness of SHERLOCK for detection of viral and bacterial pathogens (with evidence of ability to discriminate) with clinically relevant sensitivity in clinically relevant sample types. However, the authors do not provide evidence that this method provides PRECISE quantification of infectious agents in these sample types. Is there evidence that SHERLOCK could be used to track changes in viral loads over time, for example?
- Because SHERLOCK is so sensitive, it seems that contamination would be a huge problem in a clinical setting. The authors spend some time talking about using this method in areas with little resources. I would speculate that best practices in terms of sample collection and laboratory processing might be an issue in these communities where resources are so limited. How could this be addressed with this method? Are the most sensitive methods always really the best methods? What determines the most clinically useful method?
- I wondered about these as well. Regulatory questions for sure. Maureen 11:02, 14 December 2017 (PST)
- What further evidence should the authors provide to make SHERLOCK more relevant for research purposes?