To the colleagues and friends who find our paper overwhelming, and instead prefer to write and read more manageable papers with more focused messages:
I respect your perspective and your approach to science greatly, and find it invaluable, of course. Both broad and focused papers are needed in science, and we can't all write exactly the same type of paper, and we can't all be the same. Some papers/authors lay the 'landscape', other papers/authors dive deeply into the 'portraits' of specific genes and pathways.
Instead of simply dismissing our work as "TL;DR" and moving on, I encourage you to consider the value of not just releasing data, but also guiding the reader in how the data can be studied, as supporting points for detailed papers that dive into specific insights and facets.
As a genomicist, I have sought to contribute resource papers to the community across different fields (epigenomics, comparative genomics, single-cell, GWAS interpretation). Thousands of scientists have found our papers valuable through the years, and I already see the same excitement for this paper and resource from hundreds of scientists that responded overwhelmingly positively both to the resource and to the specific insights and hypotheses raised, both publicly and privately.
Following up on each of these hypotheses will take years and dedicated PhD and postdoc projects (and ultimately preclinical and therapeutic projects in industry), which are simply impossible tasks in the first landscape paper releasing the resource.
If one agrees that there is value in generating such massive datasets (and I'm happy to have this debate as well), I see three paths for publication:
(a) Simply upload the massive dataset on a website and let others figure it out (with appropriate privacy protections); or
(b) Wait 5-10 years for all the pathways and discoveries to be fully fleshed out in one giant paper or a series of papers as a 'package'; or
(c) Somewhere in the middle, namely releasing the data, providing illustrations of the types of analyses that can be carried out and the insights and tantalizing hypotheses they reveal, and carrying out initial experimental validations to provide validation support for some of our highlighted findings, with the understanding that fully pursuing any one of these insights will take many years and dedicated efforts by 'portrait' scientists.
With approach (c), which we tried to take, the large number of figures and panels in such 'landscape' papers seeks to guide the reader for how to explore the complexity of the dataset and the questions that can be answered. Each of these figures will take time to study, understand, reflect on, improve on, and hopefully each and all together will play a role in seeding ideas and insights, and pushing the field forward.
It is normal for such massive papers and resources to take days, weeks, months, and sometimes years to fully understand, and to fully extract maximum value from. It was a herculean task by the reviewers and editors, as it was of course for the authors, to go through every figure, every panel, and every result.
Of course there will be some mistakes that slipped through that process, and these will be corrected and improved up, with better methods, newer pipelines, and fresh perspectives, in science's self-correcting entreprise. We will learn from our mistakes, and from mistakes of others, and our next papers will hopefully be better, clearer, and more useful, though they will never be perfect, but hopefully sufficiently good to let others build on them and surpass them.
For those readers genuinely overwhelmed by the number of figures and panels in such papers, think of them as a good book with many chapters and many pages, not a YouTube short. Each paragraph, parenthesis, panel, supplementary figure, can hide potential hints and secrets that the authors themselves may have missed, and tools like the ChatGPT Consensus app can help elucidate, and dive deeply into specific hypotheses for hours of exploration. Ultimately, the best part of the paper begins *after* its publication, as countless students, postdocs, and group leaders dive into the findings, and weave their own stories from the threads each paper puts out.
So happy exploring and weaving, and may this paper generate more questions and hypotheses that we could have even imagined. With love to all -Manolis
Excited to share our
#MultiRegion #SingleCell dissection of
#Alzheimers out in
@Nature today
studying
#Reelin in
#RegionSpecific #NeuronalVulnerability,
#CholineMetabolism and
#PolyAmineBiosynthesis in
#CognitiveResilience,
#Astrocyte diversity,
#Thalamus-specific
#Interneurons,
#scDemon for
#Module analysis, and so much more.
Paper:
nature.com/articles/s41586-0…
News:
picower.mit.edu/news/study-a…
Website/Data:
compbio.mit.edu/ad_multiregi…
Interactive website:
compbio2.mit.edu/ad_multireg…
Cell Browser:
ad-multi-region.cells.ucsc.e…
Code:
github.com/cboix/admultiregi…
scDemon module analysis:
github.com/KellisLab/scdemon
with:
@Carles_Boix,
@MathysHansruedi,
@Leyla_Aakay, @DrLiHueiTsai,
#DavidBennett,
@RudyTanzi,
@__ben_james__, @_JoseDavila,
#KikiGalani,
@NeilBBand,
@NIHAging,
#CureAlz and so many more amazing contributors.
Truly grateful for an amazing team and collaboration
#OpenAccess #OpenScience #NeuroDegeneration