Sign in

Mike Gallagher

@geneticsmike7.bsky.social
179 followers 225 following 430 posts

Geneticist and stem cell biologist studying epigenetics and neurodegeneration at Whitehead Institute and MIT.

PostsRepliesMedia
Mike Gallagher @geneticsmike7.bsky.social · 05/10/2026
I’m confused about CiteScore - is the only difference from JIF that it has a broader definition of what counts as citable items in the denominator?
010
Mike Gallagher @geneticsmike7.bsky.social · 05/10/2026
Senior author papers only?
000
Mike Gallagher @geneticsmike7.bsky.social · 04/10/2026
And Sports Illustrated!
010
Mike Gallagher @geneticsmike7.bsky.social · 10/09/2026
I think that was somewhat specific for lncRNAs. I doubt anyone in recent yrs denied importance of tRNAs, snRNAs, microRNAs, etc. 2006 Nobel was given for RNAi!
101
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
Ginther argues that IC decisions do not differ once priority scores are controlled for, and Erosheva argues that priority scores are fully explained by the criterion scores. So the discrepancies likely occur upstream of these. Assuming these conclusions are correct, of course.
100
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
Actually many specific steps of grant review have been shown to not contribute to the discrepancy. This is important so ppl don’t waste energy looking in the wrong places.
100
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
2/ cancer researchers tend to have among the worst funding rates, and Asian PIs in the data you responded to tend to have the highest, but determining the causes of these phenomena require disentangling of all the same covariables.
100
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
1/ I disagree, whatever proportion of the effect is explained by different funding rates across topics and institutes could have multiple causes, including justified or unjustified lower fund amounts for said topics, greater supply/demand imbalance of researchers/funds in those topics, etc.
200
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
3/ intervene in any situations that are undesirable for whatever reason.
100
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
2/ very large differences in sex ratios in different profession, as well as within STEM fields. Only by teasing out confounding variables do researchers identify causal factors, and they were often not what was initially assumed. Now we are better informed and can make smarter decisions about how to
200
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
1/ Not necessarily, but a conclusion of bias specifically due to an applicants race certainly requires evidence, as do all causal claims. We can’t fix a problem if we don’t understand its causes. For example, research over the past 20yrs has significantly changed our understanding of why there are
200
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
Right, but I mean have the causal factors explaining the remaining 75% been identified?
100
Mike Gallagher @geneticsmike7.bsky.social · 09/08/2026
Has there been data demonstrating a race bias in these funding inequalities? All the papers I’ve seen have either found non-race-based explanations (eg topic) and/or couldn’t determine which variables were causal. I imagine it’s often not possible to infer one’s race based on the PI’s name.
100
Mike Gallagher @geneticsmike7.bsky.social · 07/08/2026
I spend zero time trying to predict or think about any of this, other than taking note of what happens with each fiscal year’s spending bill and how that compares with WH proposals. At minimum I’m waiting until a new administration before putting much thought into what the future might look like.
040
Mike Gallagher @geneticsmike7.bsky.social · 05/08/2026
Congrats, this is a remarkable system!
010
Mike Gallagher @geneticsmike7.bsky.social · 02/08/2026
Epigenetics is perhaps the most confused term in all of molecular biology
260
Mike Gallagher @geneticsmike7.bsky.social · 30/07/2026
3/ Importantly, increased TF binding generally increased overall accessibility, but decreases accessibility within the footprint.
010
Mike Gallagher @geneticsmike7.bsky.social · 30/07/2026
2/ as total reads could be ~30% to 3-fold different between homozygotes of the different alleles. I'm not sure if the read difference in the footprint would differ from the read difference in the overall peak.
100
Mike Gallagher @geneticsmike7.bsky.social · 30/07/2026
1/ I believe caQTL effect sizes are quite similar to eQTLs, meaning ~15-100% change in phenotype. Allele-specific effects in heterozygous samples are moderately/strongly correlated with population QTL effects. I suppose this means that peak calling could indeed be affected by variants,
121
Mike Gallagher @geneticsmike7.bsky.social · 30/07/2026
2/ and the effect of a variant is usually moderate (e.g. 30-50% more/less reads in the footprint). ATAC peaks may be more affected since they're narrower than the older DHS like in this study: www.nature.com/articles/ng....
nature.com
Large-scale identification of sequence variants influencing human transcription factor occupancy in vivo - Nature Genetics
Matthew Maurano, John Stamatoyannopoulos and colleagues identify 64,597 allelically imbalanced SNPs that influence transcription factor occupancy in vivo. Using these data, they develop a general scor...
120
Mike Gallagher @geneticsmike7.bsky.social · 30/07/2026
1/ My guess is no. DNase footprinting has been analyzed for allele-specific effects pretty broadly, and from my experience you can't even tell if there's a footprint unless you zoom in quite a bit. If a DNase/ATAC peak is ~200bp, the footprint is usually one or two dozen protected bp,
120
Mike Gallagher @geneticsmike7.bsky.social · 16/07/2026
What is this a picture of?
101
Mike Gallagher @geneticsmike7.bsky.social · 28/06/2026
1/ I agree with these points except #2 and 3. Every scientists’ work is free to be critiqued at any time. For #2, how does any correlation between specific institutions and performance imply bias? You’d have to control for lots of covariates, and perhaps compare to independent metrics.
010
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
6/ @epididymosome.bsky.social @philipcball.bsky.social @michaelzlin.bsky.social
100
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
5/ and randomness involved in many such “rankings”. This is addressed to those who abhor and reject ALL metrics/rankings, not those who only question specific ranking methodologies. Thoughts? @odedrechavi.bsky.social @richardsever.bsky.social @mbeisen.bsky.social @arjunraj.bsky.social
100
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
4/ and we don’t roll our eyes at the student who got into grad school, or the postdoc who got a TT position, just because it’s another “metric” or “ranking”, even though in reality it is! We congratulate them bc we’re happy that they achieved a goal for themselves, despite the obvious subjectivity
110
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
3/ such as QED 1%?. If the issue is that the criteria for winning are questionable, that’s fine, but some ppl seem to think that ANY ranking or metric is intrinsically antithetical to science, and therefore should be eliminated. I’m just thinking out loud - everything in life involves competition,
110
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
2/ it is common to congratulate ppl on these accomplishments, even though it implicitly acknowledges that they outcompeted X% of other candidates. If it is OK to congratulate someone who wins a fellowship as a direct result of scoring in the X%, why is it not OK to do this in other competitions,
100
Mike Gallagher @geneticsmike7.bsky.social · 26/06/2026
1/ Per the recent @qedscience.bsky.social discussions, I’m wondering how ppl think about principled consistency in evaluating selection-based criteria in science. E.g., any accomplishment that involves outcompeting other ppl (grad school admissions, publishing, grant apps, awards/fellowships, etc)
100
Mike Gallagher @geneticsmike7.bsky.social · 10/06/2026
What is the difference between them?
110
Mike Gallagher @geneticsmike7.bsky.social · 02/06/2026
And the worst offender: “decoding”
021
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
10/ haven’t read this one yet, been meaning to forever, looks very cool pubmed.ncbi.nlm.nih.gov/40369073/
pubmed.ncbi.nlm.nih.gov
Solid phase transitions as a solution to the genome folding paradox - PubMed
Ultra-long-range genomic contacts, which are key components of neuronal genome architecture<sup>1-3</sup>, constitute a biochemical enigma. This is because regulatory DNA elements make selective and s...
000
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
9/ pubmed.ncbi.nlm.nih.gov/31776509/
pubmed.ncbi.nlm.nih.gov
Chromatin structure dynamics during the mitosis-to-G1 phase transition - PubMed
Features of higher-order chromatin organization-such as A/B compartments, topologically associating domains and chromatin loops-are temporarily disrupted during mitosis<sup>1,2</sup>. Because these st...
100
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
8/ pubmed.ncbi.nlm.nih.gov/35668298/
pubmed.ncbi.nlm.nih.gov
Histone editing elucidates the functional roles of H3K27 methylation and acetylation in mammals - PubMed
Posttranslational modifications of histones (PTMs) are associated with specific chromatin and gene expression states<sup>1,2</sup>. Although studies in Drosophila melanogaster have revealed phenotypic...
100
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
7/ pubmed.ncbi.nlm.nih.gov/40966339/
pubmed.ncbi.nlm.nih.gov
Functional maps of a genomic locus reveal confinement of an enhancer by its target gene - PubMed
Genes are often activated by enhancers located at large genomic distances, and the importance of this positioning is poorly understood. By relocating promoter-reporter constructs into thousands of alt...
100
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
6/ sorry got disconnected from my flight mid-air! pubmed.ncbi.nlm.nih.gov/35951677/
pubmed.ncbi.nlm.nih.gov
Checking your browser - reCAPTCHA
110
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
5/ wrt distal regulation of genes by cis-regulatory elements and chromatin dynamics, here are a few I can think of that are insightful pubmed.ncbi.nlm.nih.gov/35418676/
pubmed.ncbi.nlm.nih.gov
230
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
4/ across cell types and states, and doesn’t have much of an effect on gene expression, at least in steady state. I think transcription and steady state RNA is buffered on so many levels that it’s very hard to disentangle causal contributions of these levels to multiomic readouts.
120
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
3/ and they data are sparse and noisy, and only get worse as you combine different measurements in the same cells. Also, we know that steady state RNA is very noisy and not a very good readout for cell state, and large-scale chromatin confirmation (typically captured by Hi-C) doesn’t change much
100
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
2/ single cell imaging experiments have been very important for understanding the stochasticity of transcriptional bursting, but I honestly can’t think of any sc-Seq approach that has moved the needle significantly. They’re often not coupled with rigorous perturbations bc they’re so expensive,
110
Mike Gallagher @geneticsmike7.bsky.social · 25/05/2026
1/ It might help, but all the big advances I’ve seen in transcriptional regulation, particularly when it comes to distal regulation by enhancers, chromosome confirmation, etc, have come from good old fashioned rigorous experiments, and none have needed any single cell sequencing.
191
Mike Gallagher @geneticsmike7.bsky.social · 23/05/2026
2/ this is not in any way a criticism of that particular paper or the authors. But the mismatch between the journal pages and funding devoted to single cell/nucleus/multiomic profiling is wildly non-scientific, IMO
000
Mike Gallagher @geneticsmike7.bsky.social · 23/05/2026
1/ That’s the goal of course, but I’ve been following these areas for many years, and I just don’t see every tweak of multiomics teaching us much that is important. It’s very impressive technically, but IMO key insights into gene regulation will not come from these approaches.
210
Mike Gallagher @geneticsmike7.bsky.social · 23/05/2026
Better question is: do we need to?
120
Mike Gallagher @geneticsmike7.bsky.social · 21/05/2026
Maybe the papers are cited more because they’re important, not necessarily because of the revisions. Maybe “big” papers are more heavily scrutinized by reviewers.
010
Mike Gallagher @geneticsmike7.bsky.social · 21/05/2026
Allman Brothers Band Eric Clapton Counting Crows Dave Matthews Band Lynyrd Skynyrd ZZ Top Vertical Horizon John Mayer Van Morrison Steely Dan
010
Mike Gallagher @geneticsmike7.bsky.social · 01/05/2026
Haven’t we?
020
Mike Gallagher @geneticsmike7.bsky.social · 19/04/2026
Affordable Care Act and Veterans Affairs. Military vets get fully funded government health care. ACA partly expanded Medicaid and provides government subsidized insurance that is bought by some ppl without employer based insurance.
010
Mike Gallagher @geneticsmike7.bsky.social · 16/04/2026
And Medicaid, ACA and VA
100
Mike Gallagher @geneticsmike7.bsky.social · 16/04/2026
Yes, typically 1.5-2X as much as OECD countries. And the outcomes are worse.
130