← pubmed.sekgen.xyz
examples
Real API responses from https://pubmed.sekgen.xyz
Six agent workflows showing what you can build with the PubMed Trends API. Responses below are real data captured from production. Where an endpoint returned empty results or CPU-limited responses, we note this transparently and continue the workflow to illustrate the full pattern. See the full API docs for endpoint details.
Use Case 1
Research Surveillance Agent
Keep your team current without reading 200 papers a week.
A biotech team working on GLP-1 agonists for cardiovascular indications needs to track the field weekly. Their agent runs on a Monday cron job: scan for spikes, check the pulse, pull key papers, and draft a Slack digest. The entire workflow costs less than a dime.
Step 1 -- What spiked this week?
GET /api/v1/hottest-weekly?category=cardiology
$0.05
{
"category": "cardiology",
"hottest": [
{ "term": "cardiac patch tissue engineering", "last_7d": 2,
"weekly_avg_30d": 1.4, "spike_ratio": 1.43 },
{ "term": "wearable ECG monitoring", "last_7d": 10,
"weekly_avg_30d": 7.46, "spike_ratio": 1.341 },
{ "term": "SGLT2 inhibitor heart failure", "last_7d": 11,
"weekly_avg_30d": 9.79, "spike_ratio": 1.124 },
{ "term": "cardiac gene therapy", "last_7d": 44,
"weekly_avg_30d": 40.79, "spike_ratio": 1.079 },
...
{ "term": "GLP-1 agonist cardiovascular", "last_7d": 24,
"weekly_avg_30d": 24.48, "spike_ratio": 0.981 }
],
"as_of": "2026-02-24"
}
The agent scans all tracked cardiology MeSH terms and ranks them by spike ratio -- last 7 days vs. the 30-day weekly average. "GLP-1 agonist cardiovascular" shows a ratio of 0.98 this week -- stable, no unusual activity. Cardiac patch tissue engineering and wearable ECG monitoring are the week's movers. Either way, the team knows.
Step 2 -- Confirm the signal with volume context
GET /api/v1/pulse?query=GLP-1+agonist+cardiovascular
$0.02
{
"query": "GLP-1 agonist cardiovascular",
"current": { "7d": 24, "30d": 105, "90d": 321 },
"year_ago": { "7d": 21, "30d": 82, "90d": 208 },
"change": { "7d": 0.143, "30d": 0.28, "90d": 0.543 },
"as_of": "2026-02-24"
}
A spike ratio can be noisy in low-volume weeks. The agent checks 7d, 30d, and 90d publication counts against the same windows a year ago. A +54% year-over-year change on the 90d window confirms the field is growing steadily, even if this particular week was quiet. The 30d change of +28% is sustained and real.
Step 3 -- Pull the papers driving the field
GET /api/v1/keypapers?query=GLP-1+agonist+cardiovascular&start_year=2020&end_year=2025
$0.03
{
"query": "GLP-1 agonist cardiovascular",
"strategies_completed": 3, "partial": false, "cached": false,
"papers": [
{ "pmid": "36876740",
"title": "Once-Weekly Semaglutide in Adults with Overweight or Obesity",
"journal": "N Engl J Med", "pub_year": 2021,
"pub_type": "clinical trial", "score": 8,
"score_breakdown": { "top_journal": 3, "review_or_trial": 2,
"relevance_rank": 2, "citation_impact": 1 } },
{ "pmid": "37840095",
"title": "Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes",
"journal": "N Engl J Med", "pub_year": 2023,
"pub_type": "clinical trial", "score": 7,
"score_breakdown": { "top_journal": 3, "review_or_trial": 2,
"relevance_rank": 1, "citation_impact": 1 } },
{ "pmid": "31535829",
"title": "Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND)",
"journal": "Lancet", "pub_year": 2019,
"pub_type": "clinical trial", "score": 6,
"score_breakdown": { "top_journal": 3, "review_or_trial": 2,
"relevance_rank": 1, "citation_impact": 0 } },
...
]
}
The agent retrieves papers scored by journal prestige, MeSH relevance, and publication type. It extracts the top 3-5 PMIDs and titles, then formats them into the Slack digest with links to PubMed. If the team uses the PubMed MCP server, the agent can also fetch full abstracts for each PMID at no additional cost.
$0.10 per weekly run. $5.20/year for continuous surveillance of one therapeutic area.
Use Case 2
Due Diligence Agent
Evaluate the science behind a biotech's pipeline in under a minute.
A VC analyst is evaluating a startup building antibody-drug conjugates (ADCs) for solid tumors. Before the partner meeting, their agent runs a quick scientific due diligence: Is the field growing? How does it compare to competing modalities? Who are the leading researchers? What does the publication landscape look like? Four calls, thirteen cents.
Step 1 -- Assess the historical trajectory
GET /api/v1/trends?query=antibody+drug+conjugate&start_year=2015&end_year=2025
$0.01
{
"query": "antibody drug conjugate",
"start_year": 2015, "end_year": 2025,
"total_results": 12002,
"data": [
{ "year": 2015, "count": 760 },
{ "year": 2016, "count": 861 },
{ "year": 2017, "count": 883 },
{ "year": 2018, "count": 956 },
{ "year": 2019, "count": 971 },
{ "year": 2020, "count": 1052 },
{ "year": 2021, "count": 1164 },
{ "year": 2022, "count": 1087 },
{ "year": 2023, "count": 1204 },
{ "year": 2024, "count": 1378 },
{ "year": 2025, "count": 1686 }
],
"analytics": {
"peak_year": 2025, "peak_count": 1686,
"cagr": 0.083, "trend_direction": "growing",
"acceleration": 0.035
}
}
The agent pulls 10 years of publication counts and computes CAGR. An 8.3% compound annual growth rate signals sustained academic and industry interest. The trend is accelerating -- 2025 hit a new peak at 1,686 papers, up 22% from 2024. The field is not a flash-in-the-pan.
Step 2 -- Benchmark against competing modalities
GET /api/v1/compare?terms=antibody+drug+conjugate,bispecific+antibodies,CAR-T+cell+therapy&start_year=2018
$0.02
{
"terms": ["antibody drug conjugate", "bispecific antibodies",
"CAR-T cell therapy"],
"start_year": 2018, "end_year": 2026,
"series": {
"antibody drug conjugate": {
"data": [
{ "year": 2018, "count": 956 },
{ "year": 2020, "count": 1052 },
{ "year": 2022, "count": 1087 },
{ "year": 2024, "count": 1378 },
{ "year": 2025, "count": 1686 }
]
},
"bispecific antibodies": {
"data": [
{ "year": 2018, "count": 378 },
{ "year": 2020, "count": 604 },
{ "year": 2022, "count": 737 },
{ "year": 2024, "count": 1226 },
{ "year": 2025, "count": 1598 }
]
},
"CAR-T cell therapy": {
"data": [
{ "year": 2018, "count": 1187 },
{ "year": 2020, "count": 1962 },
{ "year": 2022, "count": 2315 },
{ "year": 2024, "count": 3027 },
{ "year": 2025, "count": 3658 }
]
}
},
"ranking_by_growth": [
"bispecific antibodies",
"CAR-T cell therapy",
"antibody drug conjugate"
]
}
The agent compares ADCs to bispecific antibodies and CAR-T therapy -- the other major modalities competing for oncology pipeline dollars. Bispecifics are growing fastest in relative terms (378 to 1,598 papers), while CAR-T dominates in absolute volume (3,658 papers in 2025). ADCs are steady but ranked third in growth. Material context for the investment thesis.
Step 3 -- Profile the key opinion leaders
GET /api/v1/author?name=Axel+Hoos
$0.02
{
"name": "Axel Hoos",
"orcid": "0000-0001-5269-9376",
"works_count": 318,
"cited_by_count": 41761,
"h_index": 67,
"i10_index": 134,
"current_institution": {
"name": "GlaxoSmithKline (United States)",
"country": "US"
},
"top_topics": [
{ "name": "Cancer Immunotherapy and Biomarkers", "count": 201 },
{ "name": "CAR-T cell therapy research", "count": 144 },
{ "name": "Immunotherapy and Immune Responses", "count": 78 },
{ "name": "Immune cells in cancer", "count": 34 },
{ "name": "Melanoma and MAPK Pathways", "count": 30 }
]
}
The author profile reveals Hoos's expertise centers on cancer immunotherapy and CAR-T -- strong oncology credentials, but not ADC-specific. An agent flags this for the analyst: the SAB member brings oncology credibility but may lack the conjugation chemistry and linker-payload expertise core to ADC development. This is the kind of due diligence nuance that matters -- the API helps agents catch expertise mismatches, not just confirm narratives.
Step 4 -- Get the full research digest
GET /api/v1/digest?query=antibody+drug+conjugate&start_year=2023&end_year=2026
$0.08
{
"query": "antibody drug conjugate",
"papers": [
{ "pmid": "41707138",
"title": "An Antibody-Oligonucleotide Conjugate for Myotonic Dystrophy Type 1.",
"journal": "N Engl J Med", "pub_year": 2026,
"pub_type": "clinical trial", "score": 7 },
{ "pmid": "38366236",
"title": "Antibody-Drug Conjugate Overview: a State-of-the-art Manufacturing Process...",
"journal": "Pharm Res", "pub_year": 2024,
"pub_type": "review", "score": 6 },
{ "pmid": "37639687",
"title": "Exploration of the antibody-drug conjugate clinical landscape.",
"journal": "MAbs", "pub_year": 2023,
"pub_type": "review", "score": 6,
"cited_by_count": 112 },
...
],
"mesh_terms": ["Antibodies, Monoclonal", "Antineoplastic Agents",
"Immunoconjugates", "Molecular Targeted Therapy",
"Precision Medicine", ...],
"authors": [
{ "name": "Smith J", "appearances": 3 },
{ "name": "Chen L", "appearances": 2 }
],
"strategies_completed": 3
}
The agent pulls the comprehensive digest: scored key papers (with PMIDs for follow-up), the MeSH term landscape showing which sub-areas are most active, and the most frequent authors in the recent literature. Note the top result is an antibody-oligonucleotide conjugate -- a related but distinct modality. PubMed matched on shared terminology. An agent should filter or flag results that don't match the specific modality under investigation. This single response gives the analyst enough context to ask informed questions about the startup's differentiation, target selection, and competitive positioning.
$0.13 per analysis. Run it on 50 portfolio prospects and spend less than a New York coffee.
Use Case 3
Grant Writing Assistant
Data-backed Significance sections, not hand-waving.
A postdoc is writing an R01 application on CRISPR base editing for sickle cell disease. The Significance section needs to demonstrate that the field is active, growing, and ripe for the proposed work. Their agent pulls real publication data: cross-domain discoveries, emerging trends, historical trajectory, and the foundational papers that reviewers will expect to see cited.
Step 1 -- Discover cross-domain research signals
GET /api/v1/explore?topics=gene+therapy,hematology&limit=5
$0.10
{
"results": [
{ "term": "Clinical Laboratory Practices and Quality Control",
"category": "hematology",
"spike_ratio": 1.864, "cagr_3y": -0.274,
"explore_score": 1.354,
"latest_year_count": 630 },
{ "term": "Hemophilia Treatment and Research",
"category": "gene therapy",
"spike_ratio": 2.147, "cagr_3y": -0.372,
"explore_score": 1.348,
"latest_year_count": 591 },
{ "term": "Breast Cancer Treatment Studies",
"category": "gene therapy",
"spike_ratio": 2.088, "cagr_3y": -0.367,
"explore_score": 1.321,
"latest_year_count": 2480 },
...
],
"topics_scanned": ["gene therapy", "hematology"],
"as_of": "2026-02-24"
}
The agent scans OpenAlex for research signals at the intersection of gene therapy and hematology. Hemophilia treatment shows a high spike ratio (2.15x) but a 3-year CAGR of -37.2% -- a short-term surge on a declining base, possibly a major trial readout rather than sustained growth. The agent notes the discrepancy between spike and CAGR for the PI, who can use this nuance to argue for timely intervention in a field experiencing renewed attention.
Step 2 -- Find what is emerging fastest
GET /api/v1/emerging?topic=base+editing&source=openalex&min_growth_rate=0.3
$0.05
{
"topic": "base editing",
"source": "openalex",
"min_growth_rate": 0.3,
"emerging_topics": [],
"scan_timestamp": "2026-02-24T01:00:11.605Z"
}
Base editing has matured past the "emerging" threshold -- the field is now large enough that individual sub-topics no longer show 30%+ CAGR. This is actually useful context for the grant: the PI can argue that base editing has graduated from niche to mainstream, making it the right time for clinical translation work. The agent pivots to trends for the quantitative trajectory.
Step 3 -- Quantify the publication trajectory
GET /api/v1/trends?query=base+editing&start_year=2017&end_year=2025
$0.01
{
"query": "base editing",
"start_year": 2017, "end_year": 2025,
"total_results": 38632,
"data": [
{ "year": 2017, "count": 2507 },
{ "year": 2018, "count": 3143 },
{ "year": 2019, "count": 3799 },
{ "year": 2020, "count": 4085 },
{ "year": 2021, "count": 4496 },
{ "year": 2022, "count": 4554 },
{ "year": 2023, "count": 4625 },
{ "year": 2024, "count": 5070 },
{ "year": 2025, "count": 6353 }
],
"analytics": {
"peak_year": 2025, "peak_count": 6353,
"cagr": 0.123, "trend_direction": "growing"
}
}
The agent retrieves year-by-year publication counts from the first base editing papers through the present. A clear upward trajectory -- 2,507 papers in 2017 to 6,353 in 2025 -- supports the argument that "the field is at an inflection point." The agent formats this into the Significance section: "Publications on base editing grew from 2,507 in 2017 to over 6,300 in 2025, reflecting a CAGR of 12.3%."
Step 4 -- Anchor the section with foundational papers
GET /api/v1/keypapers?query=base+editing+sickle+cell&start_year=2020&end_year=2026
$0.03
{
"query": "base editing sickle cell",
"papers": [
{ "pmid": "37646679",
"title": "CRISPR-Cas9 Editing of the HBG1 and HBG2 Promoters to Treat Sickle Cell Disease.",
"journal": "N Engl J Med", "pub_year": 2023,
"pub_type": "clinical trial", "score": 9,
"cited_by_count": 138 },
{ "pmid": "33283989",
"title": "CRISPR-Cas9 Gene Editing for Sickle Cell Disease and Beta-Thalassemia.",
"journal": "N Engl J Med", "pub_year": 2021,
"pub_type": "clinical trial", "score": 9,
"cited_by_count": 1724 },
{ "pmid": "38661449",
"title": "Exagamglogene Autotemcel for Severe Sickle Cell Disease.",
"journal": "N Engl J Med", "pub_year": 2024,
"pub_type": "clinical trial", "score": 8,
"cited_by_count": 303 },
{ "pmid": "34079130",
"title": "Base editing of haematopoietic stem cells rescues sickle cell disease in mice.",
"journal": "Nature", "pub_year": 2021,
"pub_type": "research", "score": 7,
"cited_by_count": 352 },
...
]
}
The agent pulls the highest-scored papers on base editing for sickle cell disease, ranked by journal prestige and MeSH relevance. These are the citations the study section expects to see: the CRISPR Therapeutics clinical trials in NEJM, the Beam Therapeutics preclinical work in Nature. The agent extracts PMIDs and titles for the PI to incorporate directly into the Significance section.
$0.19 per grant. Less than the cost of printing the application.
Use Case 4
Drug Repurposing Scout
Scan PubMed for anomalies no one is watching yet.
A pharma innovation group runs an autonomous scanner looking for unexpected surges in MeSH term activity -- the kind of signal that might indicate a drug originally developed for one indication is showing up in an entirely different disease context. The agent runs daily, flags anomalies, drills into the fastest-growing subtopics, and assembles a digest for the team.
Step 1 -- Detect MeSH anomalies
GET /api/v1/explore?limit=10&min_z_score=2.5&days=14
$0.10
{
"results": [],
"method": "mesh_anomaly",
"baseline_days": 90,
"as_of": "2026-02-24"
}
The anomaly scan returned no results at z-score >= 2.5 for this window -- this is expected, as genuine statistical anomalies are rare. Rather than fabricate a hit, the next steps illustrate what the agent would do when following up on a detected anomaly, using GLP-1 neurodegeneration as a realistic example signal. In production, the agent logs "no anomalies detected" most days and only triggers the drill-down workflow when a real spike appears.
Step 2 -- Drill into the fastest-growing subtopics
GET /api/v1/trending-subtopics?query=GLP-1+agonist+neurodegeneration&limit=10
$0.05
{
"query": "GLP-1 agonist neurodegeneration",
"window_years": 3,
"results": [
{ "topic_name": "Diet and metabolism studies",
"counts": { "2023": 3, "2024": 5, "2025": 7 },
"cagr": 0.528, "trend_direction": "growing",
"growth_score": 0.446 },
{ "topic_name": "Diabetes Treatment and Management",
"counts": { "2023": 31, "2024": 26, "2025": 16 },
"cagr": -0.282, "trend_direction": "declining" },
{ "topic_name": "Parkinson's Disease Mechanisms and Treatments",
"counts": { "2023": 21, "2024": 17, "2025": 6 },
"cagr": -0.465, "trend_direction": "declining" }
],
"total_topics_found": 191
}
The agent drills into the subtopic structure of GLP-1 agonist neurodegeneration research. The metabolic crossover sub-area is growing at 53% CAGR, though on a small base (3 to 7 papers) -- the agent flags that small-n growth rates can be misleading. Meanwhile, the Parkinson's-specific literature is declining -- suggesting the research focus may be shifting from disease-specific trials toward broader metabolic-neurological mechanisms. The agent flags this structural shift in the report.
Step 3 -- Assemble the evidence package
GET /api/v1/digest?query=GLP-1+agonist+neurodegeneration&start_year=2023&end_year=2026
$0.08
{
"query": "GLP-1 agonist neurodegeneration",
"papers": [
{ "pmid": "41697753",
"title": "The promise of GLP-1 receptor agonists for neurodegenerative diseases.",
"journal": "J Clin Invest", "pub_year": 2026,
"pub_type": "review", "score": 8 },
{ "pmid": "39357509",
"title": "GLP-1 programs the neurovascular landscape.",
"journal": "Cell Metab", "pub_year": 2024,
"pub_type": "review", "score": 6, "cited_by_count": 33 },
{ "pmid": "38612620",
"title": "GLP-1 Receptor Agonists: A New Treatment in Parkinson's Disease.",
"journal": "Int J Mol Sci", "pub_year": 2024,
"pub_type": "review", "score": 5, "cited_by_count": 70 },
{ "pmid": "39919774",
"title": "First phase 3 trial of GLP-1 receptor agonist for neurodegeneration.",
"journal": "Lancet", "pub_year": 2025, "score": 4 },
...
],
"mesh_terms": ["Alzheimer Disease", "Glucagon-Like Peptide-1 Receptor",
"Neurodegenerative Diseases", "Neuroprotective Agents",
"Parkinson Disease", ...],
"authors": [
{ "name": "Christian Holscher", "appearances": 2 },
{ "name": "Matthias H Tschop", "appearances": 2 }
]
}
The agent pulls the full digest: scored papers with PMIDs, the MeSH landscape showing which neurological conditions are being studied with GLP-1 agonists, and the most frequent authors. A Lancet paper on the first phase 3 trial for neurodegeneration (2025) confirms the signal has reached clinical-stage validation. This package goes to the innovation team as a structured brief.
$0.23 per scan. Run it daily for a year at $84 -- less than one hour of a consultant's time.
Use Case 5
Competitive Intelligence Agent
Track what your competitors are publishing before they present it.
A pharma company developing CAR-T therapies wants to monitor how bispecific antibodies -- the competing modality -- are progressing. Their agent runs biweekly: compare the two trajectories, check for weekly spikes in competitor activity, and profile any new investigators appearing in the bispecific space.
Step 1 -- Compare publication trajectories
GET /api/v1/compare?terms=CAR-T+cell+therapy,bispecific+antibodies&start_year=2019
$0.02
{
"terms": ["CAR-T cell therapy", "bispecific antibodies"],
"start_year": 2019, "end_year": 2026,
"series": {
"CAR-T cell therapy": {
"data": [
{ "year": 2019, "count": 1463 },
{ "year": 2020, "count": 1962 },
{ "year": 2021, "count": 2413 },
{ "year": 2022, "count": 2315 },
{ "year": 2023, "count": 2455 },
{ "year": 2024, "count": 3027 },
{ "year": 2025, "count": 3658 }
]
},
"bispecific antibodies": {
"data": [
{ "year": 2019, "count": 475 },
{ "year": 2020, "count": 604 },
{ "year": 2021, "count": 780 },
{ "year": 2022, "count": 737 },
{ "year": 2023, "count": 867 },
{ "year": 2024, "count": 1226 },
{ "year": 2025, "count": 1598 }
]
}
},
"ranking_by_growth": [
"bispecific antibodies",
"CAR-T cell therapy"
]
}
The agent compares six years of publication data between the two modalities. CAR-T still dominates in absolute volume (3,658 vs 1,598 in 2025), but bispecifics are ranked higher in growth rate -- they tripled from 475 to 1,598 papers while CAR-T grew 2.5x. The gap is narrowing, and bispecifics had a 41% surge in 2024. The competitor modality is accelerating.
Step 2 -- Check for weekly spikes
GET /api/v1/hottest-weekly?topic=bispecific+antibodies&source=openalex
$0.05
{
"topic": "bispecific antibodies",
"source": "openalex",
"hottest": [
{ "term": "Monoclonal and Polyclonal Antibodies Research",
"last_7d": 1593,
"weekly_avg_30d": 12452,
"spike_ratio": 0.864 }
],
"as_of": "2026-02-24"
}
This week shows a spike ratio of 0.86 -- below average, meaning no unusual publication activity in the bispecific space. The team can relax this cycle. When this ratio exceeds 1.5, it typically indicates a major trial readout, a new FDA approval, or a high-profile publication that the team needs to know about.
Step 3 -- Profile emerging investigators
GET /api/v1/author?name=Guy+Goldberg
$0.02
{
"name": "Guy Goldberg",
"works_count": 1,
"cited_by_count": 0,
"h_index": 0,
"i10_index": 0,
"current_institution": {
"name": "Weizmann Institute of Science",
"country": "IL"
},
"top_topics": [
{ "name": "Manufacturing Process and Optimization", "count": 1 },
{ "name": "Polynomial and algebraic computation", "count": 1 }
]
}
This illustrates a critical agent task -- name disambiguation. The author endpoint returned a different "Guy Goldberg" unrelated to bispecific antibody research: top topics include manufacturing processes and polynomial algebra, with zero citations. An agent must cross-check top_topics against the research area before including a profile in competitive intelligence reports. In production, the agent would retry with additional disambiguation (e.g., institution or ORCID) to find the correct researcher.
$0.09 per report. $2.34/year for biweekly competitive intelligence on one modality pair.
Use Case 6
Quick Lookups
One call, one answer. For any agent that just needs a fact.
Not every query needs a multi-step workflow. Sometimes an agent mid-conversation just needs a single data point: what is hot in immunology this week? How has mRNA vaccine research trended? Who is this author my user mentioned? These are the building blocks -- cheap, fast, self-contained.
Example A -- What is hot this week?
GET /api/v1/hottest-weekly?category=immunology
$0.05
{
"category": "immunology",
"hottest": [
{ "term": "complement inhibitor", "last_7d": 16,
"weekly_avg_30d": 12.82, "spike_ratio": 1.248 },
{ "term": "trained immunity", "last_7d": 9,
"weekly_avg_30d": 7.46, "spike_ratio": 1.207 },
{ "term": "microbiome immunotherapy", "last_7d": 24,
"weekly_avg_30d": 23.08, "spike_ratio": 1.04 },
{ "term": "IL-17 inhibitor", "last_7d": 13,
"weekly_avg_30d": 13.52, "spike_ratio": 0.962 },
{ "term": "mRNA vaccine", "last_7d": 46,
"weekly_avg_30d": 49.42, "spike_ratio": 0.931 },
...
],
"as_of": "2026-02-24"
}
Returns the immunology MeSH terms with the highest spike ratios in the last 7 days. Complement inhibitors and trained immunity are the week's movers. An agent helping a researcher brainstorm can use this to suggest timely topics. An agent writing a newsletter can use it to open with "This week in immunology..."
Example B -- How has a field evolved?
GET /api/v1/trends?query=mRNA+vaccine&start_year=2018&end_year=2025
$0.01
{
"query": "mRNA vaccine",
"start_year": 2018, "end_year": 2025,
"total_results": 19575,
"data": [
{ "year": 2018, "count": 236 },
{ "year": 2019, "count": 265 },
{ "year": 2020, "count": 458 },
{ "year": 2021, "count": 3405 },
{ "year": 2022, "count": 5843 },
{ "year": 2023, "count": 3954 },
{ "year": 2024, "count": 2808 },
{ "year": 2025, "count": 2606 }
],
"analytics": {
"peak_year": 2022, "peak_count": 5843,
"cagr": 0.409, "trend_direction": "declining"
}
}
Year-by-year publication counts for mRNA vaccines. The data tells its own story: 236 papers in 2018, the COVID-driven explosion to 5,843 in 2022, and the post-pandemic decline to 2,606 in 2025. Still 10x the pre-pandemic baseline. Useful for any context where an agent needs to ground a claim about research momentum with real numbers.
Example C -- Who is this researcher?
GET /api/v1/author?name=Jennifer+Doudna
$0.02
{
"name": "Jennifer A. Doudna",
"orcid": "0000-0001-9161-999X",
"works_count": 667,
"cited_by_count": 115664,
"h_index": 148,
"i10_index": 397,
"current_institution": {
"name": "QB3",
"country": "US"
},
"top_topics": [
{ "name": "CRISPR and Genetic Engineering", "count": 303 },
{ "name": "RNA and protein synthesis mechanisms", "count": 284 },
{ "name": "RNA Research and Splicing", "count": 109 },
{ "name": "RNA modifications and cancer", "count": 101 },
{ "name": "Advanced biosensing and bioanalysis techniques", "count": 77 }
]
}
Returns h-index, total works, citation count, institutional affiliation, and top research topics from OpenAlex. When a user asks "tell me about Jennifer Doudna's research profile," the agent can answer with verified data instead of relying on training knowledge that may be outdated: 667 works, 115K+ citations, h-index of 148.
Example D -- Snapshot count
GET /api/v1/snapshot?query=spatial+transcriptomics&year=2025
$0.01
{
"query": "spatial transcriptomics",
"year": 2025,
"count": 3557,
"total_pubmed": 1882386,
"proportion": 0.00189
}
A single number: 3,557 papers on spatial transcriptomics were published in 2025. Useful for grounding claims ("over 3,500 papers published last year alone") or quick comparisons across fields without paying for a full trend analysis.