Skip to content

Adding zebrafish and Johri references #4

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
sooheelee wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Adding zebrafish and Johri references #4

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file modified paper/main.pdf
View file
Open in desktop
Binary file not shown.
7 changes: 4 additions & 3 deletions paper/main.tex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -190,7 +190,7 @@ \section{Evaluation Inventory}
\begin{multicols}{2}
The benchmark covers four assay types and eight task categories.
CUT\&Tag/CUT\&RUN tasks are drawn from zebrafish chromatin workflow snapshots
and span chromatin state analysis, QC, peak calling, downstream analysis,
(GSE178343) \cite{akdogan2022zebrafish} and span chromatin state analysis, QC, peak calling, downstream analysis,
genomic annotation, alignment, and visualization. ATAC-seq tasks use pediatric
B-ALL ATAC/RNA data from GSE161501 and related integrative workflow
artifacts \cite{diedrich2021allatac}, with downstream analysis dominating.
Expand Down Expand Up @@ -225,7 +225,7 @@ \section{Evaluation Inventory}
\midrule
CUT\&Tag/CUT\&RUN &
47 &
Zebrafish chromatin workflow snapshots &
Zebrafish chromatin workflow snapshots (GSE178343) &
Chromatin state (12), QC (10), peak calling (9), downstream analysis (7), genomic annotation (5), alignment (3), visualization (1) \\
\addlinespace
ATAC-seq &
Expand Down Expand Up @@ -432,7 +432,8 @@ \section{Discussion}
selected a familiar workflow default, literature-derived mechanism, or nearby
statistic that was not supported by the provided evidence. This separates
execution from judgment: an agent can operate the tools of an epigenomics
workflow and still submit the wrong biological answer.
workflow and still submit the wrong biological answer. Concurrent work on
agentic biological discovery reports the same pattern \cite{johri2026m3a}.

The benchmark has several limitations. The task inventory is not balanced
across assay families: CUT\&Tag/CUT\&RUN (47 evaluations) and methylation-seq
Expand Down
9 changes: 9 additions & 0 deletions paper/references.tex
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -74,5 +74,14 @@
\bibitem{zhang2008macs} Zhang, Y. et al. Model-based analysis of ChIP-Seq
(MACS). \textit{Genome Biology} 9, R137 (2008).

\bibitem{akdogan2022zebrafish} Akdogan-Ozdilek, B., Duval, K. L., Meng, F. W.,
Murphy, P. J. \& Goll, M. G. Identification of chromatin states during zebrafish
gastrulation using CUT\&RUN and CUT\&Tag. \textit{Developmental Dynamics} 251,
729--742 (2022). GEO: GSE178343.

\bibitem{johri2026m3a} Johri, S. et al. Evaluating agentic AI for biological
discovery in autonomous and copilot settings. \textit{bioRxiv} (2026)
doi:10.64898/2026.06.04.729919.

\end{thebibliography}
}