KKATOGEN Radar
Intelligence
This published build is older than one trading day. Treat figures as historical until the next healthy weekday refresh.
Katogen Radar · See it coming.

Every biotech catalyst on the radar — and who already owns it.

Radar tracks 420 US-listed drug developers every morning and joins what most trackers keep apart: when the data reads out, who owns the stock, how long the cash lasts, and what each means for the next name over — from SEC filings and ClinicalTrials.gov only.

New here? Radar in 30 seconds

1
Pick a section above. The grey line under the buttons tells you what each one is for before you click it. Start here is today's news; Catalyst calendar is what is coming.
2
Open “How to read this” on any panel. Every panel explains what it shows, how to read it, and what to do with it — no jargon assumed.
3
Star companies as you go. The ☆ next to any ticker adds it to My watchlist, which lives in the page address so you can bookmark it or send it to someone.

Two things worth knowing up front. Every figure links back to the filing or trial record it came from, so you can check anything you doubt. And Does it work? is Radar marking its own homework against real share-price history — including the tests it failed.

Radar watches the whole sector, free. Katogen Sentinel™ watches the names you choose and tells you when one moves. Katogen Engine™ does the deep work on one company — pipeline, valuation, risks and precedents, from 101 official sources.

What changed

Everything a tracked company filed with the SEC in the last three days, newest first.

How to read this
What it shows
Each row is one filing. The Kind tag is the form type — an 8-K is a company announcing something it is legally required to disclose; anything tagged financing is the company registering or selling shares.
How to read it
Scan the Kind column first. Financing tags next to a short runway badge are the ones that matter. Use Show to widen the window and Scope to narrow it to your watchlist.
What to do with it
Star anything you want to keep. Click open ↗ to read the filing itself — Radar never asks you to take its word for it.

Financing pressure into a binary event

Companies that have to raise money before their own trial result arrives.

How to read this
What it shows
A company with a readout inside 12 months and less than 12 months of cash at its current burn rate must finance into that event.
Definitions
The join a catalyst calendar can't make on its own. A company with a readout inside 12 months and less than 12 months of cash at its current burn has to finance into that event — which sets the terms long before the data prints. Runway = most recent reported cash & investments ÷ most recent quarterly operating cash burn. Click any row for the underlying filings.
How to read it
A company on this list is negotiating from weakness. The market knows it needs money, so the raise usually gets priced accordingly — and the terms are set long before the data prints.
What to do with it
Treat it two ways. As a risk list if you own the name — dilution is coming. As a hunting ground if you do not, because these are often the cheapest names in the sector, and they are cheap for a reason.

What the companies themselves said — last 45 days

Company announcements in their own words, straight from EDGAR — not a news feed.

How to read this
What it shows
These are 8-K filings, tagged with the item codes the company itself chose. Item 8.01 is general news, 5.02 is an officer or director leaving, 2.02 is results.
Definitions
Not a news feed. These are 8-K filings by tracked companies, straight from the EDGAR index, tagged with the item codes the company chose. An 8-K is the primary-source version of news: what the company told the SEC, when, with its own press release attached as Exhibit 99.1. Item 5.02 on a company with four months of runway reads differently to item 5.02 anywhere else — which is why these sit next to runway and ownership rather than in a separate ticker.
How to read it
The item code tells you what kind of news it is before you read a word. A company that files an 8-K on a Friday evening is usually not sharing good news.
What to do with it
Open the exhibit to read the actual press release. If a headline surprises you, check the runway and holder columns for the same company before you react.

Trading below net cash

Companies worth less than the money in their own bank account.

How to read this
What it shows
Cash and investments exceed the entire market capitalisation at today's price.
Definitions
Sometimes a dislocation; more often the market disputing the balance sheet, pricing in a wind-down, or a halted line. Read the 52-week range and volume columns before treating any of these as cheap.
How to read it
Sometimes this is a genuine dislocation. More often the market is disputing the balance sheet — pricing in the cash being burned on a programme nobody believes in, or a liability not visible here.
What to do with it
Never buy on this screen alone. Read it together with runway and burn: cheap plus a fast burn is a melting ice cube, cheap plus a long runway and a real catalyst is worth an afternoon of work.

Where the specialists moved this quarter

The biggest buys and sells by the specialist funds Radar tracks, last quarter.

How to read this
What it shows
Net change in position value across the tracked funds, valued at the quarter-end price implied by the 13F filings themselves. Top 12 each way.
Definitions
Top 12 each way.
How to read it
13F filings arrive up to 45 days after quarter end, so this is where the money was, not where it is. It is still the only honest look at what professionals actually did.
What to do with it
Use it for context, not timing. A name several specialists were adding into a catalyst is a different proposition from one they were all leaving.

Runway distribution

How many months of cash the whole sector has left, in one picture.

How to read this
What it shows
Months of cash remaining across tracked companies that report US-GAAP cash flow.
Definitions
The left two buckets are where dilution gets priced.
How to read it
The two left-hand buckets are where dilution gets priced — companies there will be raising soon.
What to do with it
If the left of this chart is getting fatter over time, the sector is heading into a financing wave and every equity story gets harder.

Catalysts per month by phase

When trial results are expected, month by month, for the next year.

How to read this
What it shows
Two kinds of date, deliberately kept apart. Registry is the primary completion date from ClinicalTrials.gov — when the last patient is measured. Guided is a date the company itself stated in a filing.
Definitions
Two kinds of date, kept visibly apart. REGISTRY is the primary completion date from ClinicalTrials.gov — when the last patient is measured on the primary endpoint, the earliest honest date a readout can happen. GUIDED is what management itself told the market, mined from the press release filed as Exhibit 99.1 to an 8-K: PDUFA dates, submission timing, topline windows. Hover any guided date to read the sentence it came from, and click through to the filing.
How to read it
A registry date is the earliest honest date data could exist; results usually follow weeks or months later. A guided date is a promise management made in writing, and is the one the market trades.
What to do with it
Look for months where several events in the same disease land together — that is when a whole group of companies reprices at once, whichever way the data goes.

Calendar

Every dated event on one timeline, nearest first.

How to read this
What it shows
Each row is one company and one dated event, with its source shown so you can tell a company promise from a registry estimate.
How to read it
Anything inside about 45 days deserves checking by hand — dates move, and the registry is often the last place to find out.
What to do with it
Star the ones you care about and they collect in My watchlist, which you can bookmark or send to someone.

Disclosed-book performance

How the funds Radar tracks have actually done — measured, not claimed.

How to read this
What it shows
The return on each manager's disclosed long US-listed equity book, holding each 13F's positions unchanged from that filing's period end to the next, chain-linked across quarters.
Definitions
What this is: the return on each manager's disclosed long US-listed equity book, holding the positions in each 13F unchanged from that filing's period end to the next, chain-linked across quarters. Every input is an SEC filing or an exchange close. What it is not: the manager's reported return — 13F shows long US equity only, at one instant, 45 days late. It excludes shorts, hedges, options, private positions, non-US listings, cash and fees, and it cannot see a trade opened and closed between two filing dates. Read it as "how has what they told the SEC they owned performed since they told us", nothing more.
How to read it
This is not the fund's real return. It ignores shorts, private positions, non-US listings, leverage, fees and anything traded between filings. It is what you could have replicated by copying their public disclosures.
What to do with it
Use it to decide whose buying you should pay attention to. A manager whose disclosed book has compounded is a more interesting signal than one whose has not.

The managers being tracked

Who these funds are, and why each one is on the list.

How to read this
What it shows
Every manager here was selected by a rule rather than by reputation — a reproducible screen over all 13F filers, described in Method & sources. "13F value" counts long US-listed equity only.
Definitions
"13F value" is the reported value of long US-listed equity positions only; it is not the manager's AUM, and it excludes private positions, non-US listings, options and debt. Biotech share and turnover are measured from the filings themselves.
How to read it
No manager was hand-picked because they are famous. That matters: a hand-picked list tells you about the person who picked it, not about the market.
What to do with it
Click any manager to see what they own. Names several of them hold at size are worth understanding before you form a view.

Consensus adds

Names the most managers were buying last quarter.

How to read this
What it shows
Companies where the greatest number of tracked managers opened or increased a position.
How to read it
Consensus cuts both ways: several specialists arriving is a real vote of confidence, and it also means the easy money may already be in the price.
What to do with it
Cross-check against the catalyst calendar. Buying ahead of a dated event is a different signal from buying after one.

Consensus reductions

Names the most managers were selling last quarter.

How to read this
What it shows
Companies where the greatest number of tracked managers trimmed or exited.
How to read it
Specialists sell for unglamorous reasons — redemptions, position sizing, a better idea elsewhere — so this is not automatically a verdict on the science.
What to do with it
Worth a hard look if you own something on this list, especially alongside a short runway.

Conviction, not just presence

Which names a manager has actually bet their year on.

How to read this
What it shows
Position weight inside a single manager's book, rather than how many managers hold it.
Definitions
"Eight managers own it" is a weaker statement than "it is 26% of one manager's book". Position count says a name is respectable; position weight says somebody is betting their year on it. Both are here, alongside current >5% holders from Schedule 13D/G, which arrive ten days after crossing rather than forty-five days after quarter end.
How to read it
"Eight managers own it" is a weaker statement than "it is 26% of one manager's book". Count says a name is respectable; weight says somebody has staked their year on it.
What to do with it
The highest-conviction positions are where a specialist believes they know something. They are also where a single bad readout does real damage to that fund — and to everything else it holds.

Crowding: how many specialists own each name

How many of the tracked funds own each company.

How to read this
What it shows
A simple count of tracked holders per name.
Definitions
A position held by many of these managers is consensus; one held by a single manager at size is a differentiated bet. Both are informative, in opposite directions.
How to read it
Many holders means consensus; one holder at size means a differentiated bet. Both are informative, in opposite directions.
What to do with it
Crowding matters most when things go wrong: a crowded name with thin trading volume falls further than the news deserves, because everyone reaches for the same exit.

Recent registration and offering activity

Who has registered shares to sell, and who has actually sold them.

How to read this
What it shows
Shelf registrations and priced takedowns filed in the last 180 days.
Definitions
An S-3 is capacity; a 424B5 is capacity actually used. Reading them next to runway tells you whether a raise was opportunistic or necessary.
How to read it
An S-3 is capacity — permission to sell later. A 424B5 is capacity actually used, meaning shares have been sold and the money is in.
What to do with it
Read this next to runway. A fresh shelf at a company with 8 months of cash tells you what is coming and roughly when.

Dilution structure — the overhang behind the runway

How much stock could arrive on the market, and through which door.

How to read this
What it shows
Registered but unsold capacity: at-the-market programmes, warrants, convertibles and options.
Definitions
Runway tells you when a company must raise. This tells you how, and how much is already registered and waiting. An ATM is capacity a company can sell into the market on any given day without another announcement — it is the quietest form of dilution and the most persistent. Warrants and convertibles sit on top. Shares outstanding is the denominator: a $150M ATM at a $90M company is a different fact from the same programme at a $3B one.
How to read it
Runway tells you when a company must raise. This tells you how, and how much is already sitting registered and waiting. An ATM can be sold into the market on any given day, without an announcement.
What to do with it
A large ATM relative to market cap caps the upside on good news, because strength is the company's opportunity to sell. Size it as a percentage of market cap, not in dollars.

Cash against burn

Every company plotted by what it has against what it spends.

How to read this
What it shows
Each dot is one company: quarterly operating cash burn across the bottom, cash and investments up the side, both on log scales. The diagonals mark 12 and 24 months of runway.
Definitions
The diagonal guides mark 12 and 24 months of runway — below the lower line is where financing stops being optional. Dots are colored by whether a Phase 2/3 readout lands inside 12 months.
How to read it
Anything below the lower diagonal has under a year of cash. Position matters more than size here — a small company burning slowly can be safer than a large one burning fast.
What to do with it
Use it to sanity-check a story. A company promising three Phase 3 programmes from a position below the 12-month line is telling you it will be raising money, whatever else it says.

Where the science is concentrated — by target

Which companies are chasing the same molecular target — the strongest link Radar draws.

How to read this
What it shows
Trial text is resolved to a drug target from what the sponsor writes, not from the disease the drug is aimed at, so an amylin analogue and an oral GLP-1 do not get treated as the same thing.
Definitions
The stronger of the two linkages. Trial text is resolved to a molecular target and a modality from what the sponsor writes, not from the disease the drug is pointed at — so an amylin analogue never reads across to an oral GLP-1 just because both say "obesity".
How to read it
Companies sharing a target share the same underlying biology risk. If the mechanism works, several of them look better at once; if it fails, several look worse.
What to do with it
Before any dated readout, look at who else is on that target. Their prices move on someone else's data, and that is usually where the mispricing sits.

Where the money is concentrated — by indication

Which companies are chasing the same disease — a weaker link, worth reading second.

How to read this
What it shows
Companies grouped by the condition their trials treat, regardless of mechanism.
Definitions
The weaker linkage, and the one worth reading second. Two companies in the same indication compete for the same patients and the same payer budget even when their mechanisms are unrelated — so this is a commercial read, not a scientific one.
How to read it
Two companies in one indication compete for the same patients and the same payer budget even when their science is unrelated. That is a commercial link, not a scientific one.
What to do with it
Use it to judge how crowded a market is getting. Being fourth into an indication is a very different commercial proposition from being first.

Readouts that price the class

Upcoming results that will move more than one company.

How to read this
What it shows
Every readout inside 12 months that has at least one tracked peer, with peers split by whether they share the target or only the indication.
Definitions
Peers linked by TARGET share the molecule's mechanism and read across hardest; peers linked by INDICATION compete commercially but may work completely differently. Direction is never implied — a competitor's win can validate a mechanism or take the market.
How to read it
Target-linked peers read across hardest, because the biology is shared. Indication-linked peers compete commercially and move less.
What to do with it
These dates are worth putting in a diary even for companies you do not own, because they reprice names you might.

Trials that stopped — and who else is in that indication

Trials that were terminated, suspended or withdrawn — and who else is exposed.

How to read this
What it shows
Phase 2/3 industry studies that stopped, with the sponsor's own stated reason and the tracked companies working in the same area.
Definitions
Read the reason carefully: "sponsor decision" and "strategic reprioritisation" are portfolio calls, while a safety or futility stop is a signal about the target, and that one travels to everyone working on it.
How to read it
Read the stated reason carefully. "Sponsor decision" and "strategic reprioritisation" are doing a lot of work in these sentences and often mean the data was not good enough to continue funding.
What to do with it
A stop is a fact, not an opinion — unlike a rumour, it is filed. Check whether anything you own shares the target with the trial that stopped.

Financing read-through

What one company's fundraise tells you about what its peers will get.

How to read this
What it shows
Each raise is paired with tracked companies in the same indication, ordered by who has the least cash left.
Definitions
When a company prices a deal, it sets the clearing price for everyone comparable. Each raise below is paired with the tracked names in the same indication, ordered by who has the least runway — the ones most likely to have to follow.
How to read it
When a company prices a deal, it sets the clearing price for everyone comparable. A deal done at a steep discount tells every similar company what its own raise will cost.
What to do with it
If you own something in that group with a short runway, a bad print from a peer has just told you the terms of your own next financing.

Registry change detection

Quiet edits companies made to their own trial records.

How to read this
What it shows
ClinicalTrials.gov keeps every version of every record. This compares them and surfaces what changed: completion dates moved, enrolment targets cut, primary endpoints amended.
Definitions
None of that is visible on any interface, so a trial running eighteen months late looks exactly like one running to plan. This compares what the sponsor registered on day one against the record as it stands today. Long-term safety and extension studies are marked LTE — a decade of slippage there is normal and means nothing.
How to read it
None of these changes is announced. A primary endpoint amended after a study opened is the single most informative edit here, and it is almost never in a press release.
What to do with it
Treat a slipped date plus a changed endpoint as a serious warning about a trial, regardless of how confident the company sounds.

Prior probability of success

A starting estimate of whether a trial will work, before anyone has seen the data.

How to read this
What it shows
It begins with published success rates for that therapeutic area and phase, then adjusts up or down for what the registered protocol actually says — randomised, blinded, size, endpoint type, and whether anything was changed after the study opened.
Definitions
This is a prior, not a prediction. It starts from published phase-transition success rates for the therapeutic area — industry benchmarks from the clinical-development literature, not Radar's own estimates — and adjusts using only what the sponsor registered: randomisation, masking, enrolment, endpoint hardness, and whether the record has been amended since. It knows nothing about the drug. Its job is to rank six hundred rows so you know where to spend an afternoon, and every component of every adjustment is shown so you can disagree with it precisely.
How to read it
This is a prior, not a prediction. It knows nothing about the molecule, the preclinical work or the people running the trial. Every adjustment is itemised so you can disagree with it line by line.
What to do with it
Use it to spot where the market's confidence and the base rate disagree. A widely loved programme with a poor design score is exactly where being early is worth something.

Does any of this actually move the price?

The board grading its own homework against real share-price history.

How to read this
What it shows
Returns are abnormal — the company's move minus XBI's over the same days — because a 6% day in a week the whole sector rose 6% is not information.
Definitions
Every other tab is descriptive. This one measures what happened to the share price around the events Radar tracks, so the panels can be judged rather than believed. Returns are abnormal — the name's move minus XBI's over the same window — because a 6% day in a week the whole sector moved 6% is not information. Windows are short and samples are small; treat these as orders of magnitude, not as parameters.
How to read it
Windows are short and samples are small. Treat these as orders of magnitude rather than as numbers to trade against.
What to do with it
Read this before you trust any other section. Anything Radar could not demonstrate is labelled as such, including the tests that failed.

8-K reaction by item code

Priced offerings

Insider buying and what followed

Does a competitor's failure travel? — testing the cross-read engine

A direct test of the knock-on-effects section's central claim.

How to read this
What it shows
Take every stopped Phase 2/3 trial, then measure what the sponsor and its linked peers did over the following five sessions.
Definitions
The Cross-reads tab asserts that a stopped trial reads through to companies working on the same target. That is a claim, and it is testable: take every stopped Phase 2/3 trial, and measure what the sponsor and its linked peers did over the following five sessions. If the target link is real it should carry more signal than the indication link.
How to read it
If the target link is real it should carry more signal than the indication link. That is a falsifiable claim, and this is the result.
What to do with it
If the effect here is small, size your read-across trades accordingly. A real but small edge is still an edge — it is just not a reason for a large position.

Watchlist

The companies you starred, gathered in one place.

How to read this
What it shows
Star any company anywhere on Radar and it collects here.
Definitions
The list lives in the page URL — — so it survives a refresh and can be sent to someone else as a link. Nothing is stored on a server and nothing is stored in your browser.
How to read it
The list lives in the page address itself. Nothing is stored on any server and no account is needed.
What to do with it
Bookmark the page to keep your list, or send the address to a colleague and they will open the same watchlist you are looking at.

Screener

Every tracked company in one sortable, filterable table.

How to read this
What it shows
Every drug developer held by at least one tracked fund. Market cap is today's price × reported shares outstanding, cross-checked against the exchange figure.
Definitions
Every drug developer held by at least one tracked fund. Sort any column; search matches ticker and company name. Market cap is today's price × reported shares outstanding, cross-checked against last reported public float; where the share basis doesn't match the traded line (ADS versus ordinary, or a split), the exchange's own published figure is used and marked EXCH.
How to read it
Sort any column by clicking its heading. Search matches both ticker and company name.
What to do with it
This is the place to build a shortlist: sort by runway to find financing pressure, by holders to find crowding, then star what survives and work through it in My watchlist.

How this is built

Sources

  • Positions — Form 13F-HR information tables, current quarter against prior quarter, for every manager passing the screen described below.
  • Manager selection — SEC bulk Form 13F data sets, plus the EDGAR company list for SIC 2833–2836 and 8731 used to classify holdings.
  • Company identity — SEC EDGAR submissions API; issuer names on 13F tables are resolved to CIK and current ticker, and unresolvable issuers are excluded rather than guessed.
  • Balance sheet — SEC XBRL company-facts frames: cash and equivalents, short and long-term investments, operating cash flow, R&D expense, shares outstanding, public float.
  • Financing — EDGAR filing index for S-1/S-3/F-1/F-3 registrations and 424B priced takedowns.
  • Catalysts — ClinicalTrials.gov v2 API, industry-sponsored interventional Phase 2 and Phase 3 studies with a primary completion date in the window.
  • Trial design and registry history — ClinicalTrials.gov study records and their full version history. Every record's first registered version is compared against today's to detect date slippage, enrolment cuts and primary-endpoint amendments.
  • Insider transactions — SEC Form 4, open-market codes only, split between individual executives and fund-affiliated directors.
  • Ownership above 5% — Schedule 13D and 13G cover pages.
  • Dilution structure — XBRL warrant, option and convertible-debt tags, plus 424B and S-3 prospectus cover pages for ATM and shelf capacity.
  • Prices — end-of-day / last-sale quotes through a provider chain: Katogen Engine market_quotes_eod first, then Finnhub, then Polygon, then keyless exchange feeds. Each price carries its own source and timestamp; the source actually used for this build is shown in the header and on every price tooltip.

Refresh cadence

  • EDGAR filing index — hourly; new S-3 and 424B5 filings appear the same day.
  • ClinicalTrials.gov — daily.
  • XBRL fundamentals — daily, refreshed materially at each 10-Q.
  • 13F positions — quarterly, within hours of the 45-day deadline.
  • Prices — end of day, with a one-hour cache; the same call runs intraday if you want it to.

What this is not

  • Not a PDUFA calendar. Primary completion date is the earliest defensible readout date from the registry. Company-guided readout windows and PDUFA dates live in 8-Ks and press releases and are added per-name on request.
  • Not a real-time terminal. Prices are end-of-day / last-sale, not streaming, and are stamped with the time they were taken. Where a quote can't be resolved Radar falls back to the price implied by the 13F filing itself at quarter end and says so.
  • The prior probability is a prior. It comes from published base rates plus registered study design. It knows nothing about the molecule, the preclinical package, the investigators or the competitive set. It is there to rank a long calendar, not to price a position.
  • The event study is descriptive. Short windows, small samples, one market regime, no correction for multiple comparisons. It shows what happened, not what will.
  • Not investment advice. It is a monitoring surface. Nothing here is a recommendation.
  • 13F is a lagged, partial picture. It shows long US-listed equity only, 45 days after quarter end, and says nothing about shorts, hedges or private positions.

Data-integrity rules applied

  • Cash is capped at reported total assets; where a company tags a combined cash-and-investments total, that figure is used rather than summed components, to avoid double counting.
  • Implied market cap is cross-checked against last reported public float and withheld where the ratio implies an ADS/ordinary share or split mismatch.
  • Market cap is live price × reported share count, but where that implies a wildly different figure to the last public float — the ADS-versus-ordinary-share trap — the exchange's own published market cap is used instead, and the substitution is disclosed on the figure.
  • Foreign private issuers reporting under IFRS are marked; US-GAAP-only fields are left blank for them rather than shown as zero.
  • Issuer-to-ticker matching requires a first-token match plus high string similarity; near-misses are dropped, not fuzzily assigned.
  • Long-term safety and extension studies are marked LTE and excluded from slippage statistics by default — a decade of drift in an open-label extension is normal and carries no information.
  • Every component of the prior is displayed with its sign and size. Nothing is weighted invisibly.

How the universe was selected

The managers — screened, not chosen

No manager is on Radar because it is well known. The list is the output of a screen run over the SEC's bulk Form 13F data set — every 13F filer in the market, 8,672 of them — with each holding classified as a drug developer or not against the 2,900 SEC registrants in SIC 2833–2836 and 8731.

A manager is included when all five hold:

  1. ≥ 60% of reported 13F value sits in drug developers — the mandate is the sector, not an index weight.
  2. ≥ $1.0B of biotech value, so a position is large enough to signal.
  3. ≥ 20 reported positions — this removes foundations, endowments and corporates that file a 13F for a single strategic stake.
  4. Two consecutive 13F-HR filings, so positions can be diffed quarter to quarter.
  5. Not itself an operating registrant — no 10-K, 20-F or 40-F. This removes pharma companies holding stakes in other pharma companies.

Every threshold is a parameter in screen_rank.py; lowering the dollar floor to $500M adds roughly sixteen more managers. The screen is re-run each quarter, so a manager that stops being a specialist drops out on its own.

The companies — derived, not chosen

No company was picked by hand. The universe is whatever those managers actually own, filtered to drug developers:

  1. Take every issuer on the screened managers' latest 13F information tables — — distinct issuers.
  2. Resolve each to a CIK and current ticker with a strict matcher; unresolvable issuers, typically acquired or delisted names, are dropped rather than guessed.
  3. Keep only SIC codes 2833–2836 and 8731 — pharmaceutical preparations, biological products and commercial biological research. This removes device, diagnostics, payer, hospital and services names the same funds hold.
  4. What survives is — drug developers. Every one is there because a screened manager owns it: this is the investable universe as the specialists have defined it with their own capital, not as anyone here has curated it.

What this selection is good and bad at

  • Good — high signal density; every name has at least one informed holder, and crowding is measurable.
  • Blind spot — a company no tracked fund owns is invisible here, including recent IPOs and names the specialists have wholly exited. Roughly 81% of the Phase 2/3 trials in the window are run by sponsors outside this universe.
  • Fix — the universe is one list in build_data.py. Widening it to all listed biotechs, or to an internal watchlist, is a configuration change, not a rebuild.

Coverage

See it first · Decide right · Get it done

This board tells you where to look. Engine tells you what to do about it.

Katogen Engine™ takes any name on this page and runs it against 101 official sources — FDA, EMA, MHRA, PMDA, CMS, USPTO, EDGAR, ClinicalTrials.gov — returning an Executive Snapshot, Pipeline Deep-Dive, Risks & Catalysts, Strategic Implications or Investment Thesis, cited to primary documents. Engine prioritises official sources and trial validation over speed.

Continue with your work email — no password, three reports free every month. Katogen Sentinel™ then watches the names you care about and tells you when something moves.

Start with Engine — 3 free reports Intelligence · Strategy · Operations