Chem2Cure Research and Therapeutics
Preclinical drug discovery

Starting a discovery culture where none has yet taken root

Bangladesh has strong pharma manufacturing capacity, but original small-molecule discovery remains largely absent. Chem2Cure aims to build the scientific habits, teams, and workflows needed to change that trajectory.

Penicillin G, the first antibiotic made at scale.
Why this is hard

A decade from an idea to a medicine

A new medicine passes through nine stages before a patient can take it, and more than a decade usually passes along the way. Most candidates fail, and each stage needs different people, instruments, and judgement. The point of showing it is not to discourage anyone: it is to be clear about which part of this chain Bangladesh is currently equipped for, and which part we are trying to build.

Target hypothesis Virtual screening, high-throughput assay development
Often decades, usually in academia
Target and hit validation Mechanism of action established
Hit ~1 year
Hit to lead Series built around the hit
Lead ~1.5 years
Lead optimisation Potency, selectivity, ADME, safety
Candidate ~1.5 years
Preclinical Animal efficacy and toxicology
IND ~1 year
Phase I Safety, tens of healthy volunteers
~1.5 years
Phase II Efficacy, hundreds of patients
~2.5 years
Phase III Confirmation, thousands of patients
~2.5 years
Submission to launch Regulatory review and approval
Approved medicine ~1.5 years

Scroll sideways to see all nine stages

Where Bangladesh sits today

The country's pharmaceutical industry is genuinely strong, and supplies most of the domestic market. But that strength lies downstream of everything above: it begins once a molecule already exists, has already been proven, and its patent has expired. None of the nine stages above happens here at any scale.

API manufacture Making the active ingredient at scale
Formulation Tablets, capsules, injectables
Quality control Analytical release testing
Packaging and supply Distribution and export
The gap we are working on

Making and formulating a known molecule is manufacturing. Deciding which molecule to make in the first place is discovery, and it needs chemists trained to read a target, design a series, and judge the data. That training is what Chem2Cure exists to provide.

See the training tracks

Timelines are typical industry figures and vary widely by therapeutic area. Most programmes never reach the next stage at all.

First principles

Discovery is not one tool or one assay. It is a disciplined loop between chemistry, biology, computation, pharmacology, safety, and decision making.

Where the time goes

AI compresses the front of that chain, not the whole of it

Look again at the first two stages above. Target work and hit finding are where programmes lose years, because the only way to know whether a molecule works has traditionally been to make it and test it. That is the part computation changes. Phase I to III still take their years, whatever anyone claims.

From a library of millions to a handful worth making

Screening and generative design narrow the field computationally before anything is weighed out, and retrosynthesis checks that what survives can actually be built from material you can buy. The compounds that reach the bench are then the ones most likely to matter, which is how months of synthesis get saved.

An AI-aided discovery workflow: a compound library feeds virtual screening, generative design proposes new analogues around a scaffold, retrosynthesis breaks the chosen molecule into purchasable starting materials, and the shortlist is then tested in a biological assay to identify a hit compound.
Virtual screeningScore a library computationally before anything is weighed out
Generative designPropose analogues around a scaffold, with the chemist choosing which are sensible
RetrosynthesisCheck the molecule can be made, from building blocks you can buy
Biological assayTest the shortlist, and confirm a hit in the laboratory

Nothing here removes the experiment. A predicted score is a hypothesis, and the assay is still what decides. What changes is how many hypotheses you have to pay for before one works.

Discovery operating model

A practical preclinical pipeline for Bangladesh

01

Target and disease logic

Prioritize diseases, targets, and mechanisms where Bangladesh can build a meaningful research foothold.

02

Hit finding

Use literature, patents, virtual screening, fragment thinking, and existing chemical matter to identify starting points.

03

Synthesis and SAR

Design analogs, build focused libraries, interpret activity, and connect molecular changes to biological response.

04

ADME and safety

Introduce solubility, permeability, stability, selectivity, and early safety thinking before expensive failure.

05

Preclinical readiness

Prepare the evidence package needed for partnerships, grants, and translational next steps.

AI-aided discovery workflow

Input: target structure, known ligands, disease context

Model: structure prediction, docking, affinity ranking, property filters

Chemist review: synthetic feasibility, novelty, SAR hypothesis

Output: prioritized molecules for purchase, synthesis, or collaboration

AI readiness

AI should help chemists think better, not replace scientific judgment

Chem2Cure introduces AI as an integrated research workflow: structure prediction, docking, library triage, property prediction, and literature intelligence. Every output still needs human medicinal chemistry judgment.

  • Target and protein-structure preparation
  • Virtual screening and prioritization
  • Chemist-reviewed shortlist generation
  • Cost-aware experimental planning
Research themes

Areas where Bangladesh can build capability

Neglected and infectious diseases

Dengue, emerging viral threats, and regionally relevant diseases where local need is high.

Rare and genetic diseases

Capacity building around diagnosis, therapy discovery, and patient-centered translational research.

Fine and specialty chemicals

Skill transfer toward fragrances, food chemicals, agrochemicals, and higher-value chemical industries.

CRO and CDMO readiness

Training chemists for contract research, analytical documentation, and global partnership standards.