Computational biologist at Broad Institute: Cell Painting profiling, encoder benchmarks, AI-agent science evaluation
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
44
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Not scored: the verified publication-count data this axis needs is not yet available in this profile's record. Output would measure total peer-reviewed output, its distribution across venues and years, and authorship roles over time.
Impact—
Not scored: citation-based influence measures are not yet available for this record. Impact would capture citation counts, the citation trajectory over time, and the uptake of this researcher's methods by other groups.
Focus100/100
Focus scores 100, the top of the scale. The topic distribution is strongly concentrated: Cell Image Analysis Techniques leads with 94 works, ahead of single-cell and spatial transcriptomics (48) and computational drug discovery methods (39); the one field-tagged catalog paper sits in computational biology / image-based profiling.
Collaboration54/100
Collaboration scores 54, a moderate level. The record lists 14 distinct coauthors in the recent window and 12 recurring coauthors, including Tim Treis, Nikita Moshkov, Johan Fredin Haslum, Fabian J. Theis, Sangeeta N. Bhatia, and Ryan Tewhey, spanning single-cell computation and imaging-engineering groups; both catalog papers are middle-author contributions.
Activity100/100
Activity scores 100. The record shows 72 works in the most recent three-year window, with year-by-year counts climbing from 12 in 2022 to 23 in 2023 and a peak of 33 in 2024; 25 works are recorded for 2025 and 14 already for 2026, indicating sustained, rising output.
Reproducibility55/100
Reproducibility scores 55, based on only one evaluated paper (reproMean 55, 1 paper). The reviewed Cell Painting encoder-evaluation preprint carries a medium reproducibility readiness; the review noted convincing benchmark design but single-run results and unreleased artifacts, so this score should not be generalized across the portfolio.
Research summary
Shantanu Singh is a computational biologist at the Broad Institute (US) whose work centers on image-based profiling and computational methods for cell biology. The OpenAlex-sourced record behind this profile lists an h-index of 44 and an i10-index of 79, with topics concentrated in Cell Image Analysis Techniques (94 works), single-cell and spatial transcriptomics (48), computational drug discovery methods (39), digital imaging for blood diseases (27), and cellular mechanics and interactions (14). Output has accelerated in recent years: 72 works in the most recent three-year window, with year-by-year counts rising from 12 in 2022 to 23 in 2023, a peak of 33 in 2024, 25 in 2025, and 14 already in 2026. The activity and focus dimensions both reach 100, reflecting this concentration. Two 2026 arXiv preprints appear in the Evolve catalog: a paired-view framework for Cell Painting encoder evaluation (CP-BG-Bench), reviewed as a convincing design with single-run results and unreleased artifacts, and a study of science sandboxes that measure the scientific capability of AI agents. Both papers list this researcher in a middle, non-corresponding position. Collaboration scores 54, with 14 distinct coauthors in the recent window and a reproducibility score of 55 based on one evaluated paper.
Collaboration pattern
The record lists 14 distinct coauthors in the recent window plus 12 recurring ones: single-cell computation (Fabian J. Theis, Nikita Moshkov, Johan Fredin Haslum, Tim Treis), imaging and disease engineering (Sangeeta N. Bhatia, Arya S. Rao), and genomics and drug discovery (Ryan Tewhey, Steven K. Reilly, Sager J. Gosai, Kenneth B. Hsu, Rodrigo I. Castro, Yasha Ektefaie). Both reviewed 2026 papers are middle-authored, non-corresponding contributions, consistent with a contributing-specialist role.
Research keywords
Highlights
Paired-view benchmark shows Cell Painting encoder rankings are protocol-dependent
Science sandboxes as a measure of AI agents' scientific capability
A second 2026 preprint in the catalog examines science sandboxes that measure the scientific capability of AI agents, extending this profile's record toward evaluation methodology for AI in science. It has not yet received an Evolve verdict, keywords, or reproducibility assessment.
Recent output surge led by a 2024 peak
The OpenAlex year-by-year record rises from 12 works in 2022 to 23 in 2023 and a peak of 33 in 2024, with 25 works in 2025 and 14 already in 2026; the most recent three-year window holds 72 works, driving activity and focus scores of 100.
Timeline
- 2017Output reaches 11 works in the year-by-year record, the first clear step up from 6 in 2016.
- 202012 works are recorded, continuing a steady cadence into the early 2020s.
- 2023Annual output rises to 23 works, nearly doubling from 12 in 2022.
- 2024Peak year in the record with 33 works.
- 2026Two preprints enter the Evolve catalog — the CP-BG-Bench Cell Painting encoder-evaluation framework and a study of science sandboxes for AI agents — with 14 works already recorded this year.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric metrics (h-index 44, i10-index 79, topic counts, year-by-year counts) come from an OpenAlex-sourced snapshot; the output and impact dimensions lack underlying data and are not scored.
- The reproducibility score (55) rests on a single evaluated paper (reproPapers: 1) and should not be read as a portfolio-level assessment.
- Only two 2026 arXiv preprints are in the Evolve review catalog; one has no verdict, keywords, or reproducibility assessment yet, and neither paper is a corresponding-author contribution.
- Identity is name-resolved from public bibliographic data (OpenAlex, ORCID 0000-0003-3150-3025); the name Shantanu Singh is shared by multiple researchers, so scope is limited by institutional affiliation.