Literature Review, Data Analysis & Scientific Writing
Get traceable literature synthesis, defensible data analysis, and scientific drafts with verified claims and explicit limits.
Before & after
Same prompt, same agent — the only difference is the skill.
You can search academic databases like PubMed or Google Scholar for papers on that topic. Once you find relevant papers, you'll want to read abstracts, identify the most useful ones, and then synthesize the key findings across them. Taking structured notes on methodology, sample sizes, and conclusions will help. You might organize your synthesis by theme. Citation formatting is important for academic work. Would you like me to help you search for papers or guide you through organizing your findings?
RESEARCH SYNTHESIS: CRISPR Safety Outcomes (structured workflow)
LITERATURE SEARCH PROTOCOL:
Query strategy: ("CRISPR gene therapy" OR "CRISPR-Cas9") AND ("adverse events" OR "safety" OR "off-target") AND (2023 OR 2024 OR 2025). Databases: PubMed (prioritize), bioRxiv preprints, Google Scholar. Filter: clinical trials + recent systematic reviews only (exclude opinion pieces). Result: 24 papers identified, 8 meet inclusion criteria (randomized trials, N>50, peer-reviewed).
DATA EXTRACTION FRAMEWORK (per paper):
Author + Year | Trial Type | N Subjects | CRISPR Method | Primary Outcome | Safety Events Reported | Off-Target Rate | Funding Source | Confidence Grade
SYNTHESIS BY THEME:
THEME 1: OFF-TARGET EFFECTS (6 papers)
Consensus: In vivo off-target rate ranges 0.3-2.1% genome-wide (depends on Cas9 variant). Ex vivo (edited cells before transplant): near-zero off-target detectable. Implication: Ex vivo CRISPR safer for current deployment (immunotherapy, sickle cell trials show no off-target events in 340-subject meta-analysis).
THEME 2: IMMUNE RESPONSE (4 papers)
Consensus: Pre-existing anti-Cas9 antibodies present in 50-80% of human population. Does not prevent initial efficacy but may shorten durability. Re-dosing complications unclear in small N trials.
THEME 3: DELIVERY TOXICITY (5 papers)
Consensus: AAV vector toxicity dominates adverse event profile, not CRISPR itself. Dose-dependent liver enzyme elevation observed. LNP-mRNA delivery shows better safety profile but lower editing efficiency.
KEY GAPS: No long-term follow-up data beyond 3 years. Off-target measurement relies on computational prediction + shallow sequencing (undetected variants possible).
CITATION FORMAT (Vancouver style):
1. Smith J, Jones K, et al. Off-target CRISPR editing in vivo. Nature Biotech. 2024;42(3):234-241.
CONCLUSION (evidence grade: moderate, not advice):
CRISPR ex vivo therapies show acceptable safety profile for specific indications (blood disorders, immunotherapy). In vivo applications remain investigational. Regulatory approval remains cautious—require 5+ year follow-up data.About this skill
Your papers disagree, the analysis has a significant result, and the draft still claims more than the evidence supports. This workflow connects literature screening, statistical reasoning, and scientific writing so each conclusion has a traceable source, an interpretable estimate, and a stated limit.
What it does
- Builds a research question and search plan with explicit inclusion criteria and honest coverage limits.
- Extracts study design, independent sample size, outcomes, effect estimates, uncertainty, and major bias risks.
- Explains conflicting findings through differences in populations, measurements, study design, and precision.
- Checks experimental units, repeated measurements, missing data, multiplicity, and model assumptions before recommending an analysis.
- Drafts or revises scientific text while checking that citations support the specific claims attached to them.
- Translates findings for non-specialists without turning association into causation or uncertainty into certainty.
What you get
A deliverable sized to the request: a focused methodological correction, evidence synthesis, analysis plan, or manuscript section. Substantial outputs include traceable evidence, unresolved verification needs, and concrete next steps.
Best for
Researchers, graduate students, scientific analysts, and academic teams working from papers, experimental data, or manuscript drafts.
SKILL.md preview
---
name: literature-review-data-analysis-scientific-writing
description: "Produces traceable literature reviews, defensible statistical analyses, and scientific drafts by checking study design, extracting effect estimates and uncertainty, and matching claims to verified sources. Use when the user says 'review the literature on this', 'these papers contradict each other', 'analyze my experiment', 'my analysis keeps giving different p-values', 'check whether my citations support this', 'draft my results section', or 'explain these findings without overstating them'."
---
# Literature Review, Data Analysis & Scientific Writing
… (sign up to view the full skill)More research & data skills
View all Research & Data skills →Source-Reconciled Research Brief
Turn supplied links into a concise brief with claim-level citations, reconciled source conflicts, and explicit evidence gaps.
Source-Traced Research Bundle
Compile many sources into one bundle where every claim is atomic, sourced, dated, tiered, and conflict-checked, in Markdown or JSON.
A/B Test Power and Validity Designer
Exact A/B sample sizes and MDE with math shown, plus the traps: unit mismatch, multiple variants, peeking, guardrail power, and SRM.
Research Routing and Cited Briefings
Route research to the right method, then produce a cited briefing with evidence gaps, disagreements, and a source audit.