From sequencing data to ranked targets.
- Bulk RNA-seq, whole-genome, whole-exome, copy-number and single-cell pipelines
Ranked, evidence-linked targets, handed straight to Silico X plore
45+ guided modules five analysis approaches
AI-driven bioinformatics and drug discovery
From genome to molecule.
Artificial intelligence and physics-based tools that turn sequencing data into prioritised targets, and targets into computationally verified candidate molecules. No coding required.
01OmicsX plore Target discovery
AI-powered analysis of multi-omics data to identify and prioritise novel therapeutic targets.
02SilicoX plore Molecular design
AI-driven design and optimisation of molecules with predicted high affinity and selectivity.
-
Incubated at
A Department of Biotechnology bio-innovation hub in the Bangalore Life Science Cluster.
-
Model partner
BoltzBoltz models, integrated into Silico X plore through our partnership. -
Peer reviewed
30+
Publications from the group, across a wide range of journals.
Our technology
Two technologies. One connected workflow.
Each stands on its own. Used together, the output of one is the input of the other.
From target to ranked candidates.
- Generate, dock, simulate and rank candidates in one browser session
- Several docking engines, molecular dynamics, free energy and DFT
- Boltz structure and affinity models, integrated through our partnership
31+ modules seven functional areas
How it works
Discovery is a loop, not a line.
AI proposes, physics verifies, and your experiments make the next cycle better.
- 01
Omics X ploreFind
Evidence-linked targets, ranked from your sequencing data.
- 02
Silico X ploreDesign
Generative AI proposes molecules; docking and simulation verify them.
- 03Your laboratory
Test
You assay the nominated candidates.
- 04Both technologies
Learn
Assay results retrain the models for the next cycle.
Results
Results on real discovery campaigns.
Each technology has been applied end to end.
752,436→4Molecules generated, then nominated
NDM-1: a resistance enzyme with no clinical inhibitor.
A ten-stage cascade, run in one session, took a generative campaign against a carbapenem-resistance enzyme down to four candidates, checked by docking, molecular dynamics and free energy.
Mangal P, Shinde O, Chikhale S, Bhowmick S, Islam MA. Journal of Molecular Graphics and Modelling 149 (2026) 109566.
SilicoSAR · OncologyPublished · 2026
36,000→8Natural products screened, then carried to free energy
EGFR: our machine-learning QSAR module, published and applied.
SilicoSAR builds and validates a potency model from known actives, then screens a natural-product library through ADMET, docking and 200 ns dynamics in the same session. It has now been through peer review.
Islam MA, Bhowmick S, Chikhale S. Computational Biology and Chemistry 126, 109467.
3,464Differentially expressed features
IPF: 39 lung transcriptomes, reads to enrichment.
One guided run took raw reads from patients and control donors through to functional enrichment. Extracellular matrix organisation dominated the results.
Battula M, Mangal P, Shinde O, Chikhale S, Bhowmick S, Islam MA. Manuscript, not yet published.
Why SilicoScientia
Built by scientists for scientists.
Each module comes from our own discovery programmes and is tested in practice before it reaches you.
No-code by design
One guided form replaces Linux, installation, scripting and job schedulers. Results return as publication-ready plots and tables.
One chemical representation
Established engines, openly cited
Tools such as AutoDock Vina, GROMACS and RDKit run unmodified under their own licences, and are cited on every result.
Reproducible by construction
Every job records its parameters, engine versions and logs. An independent re-run of our IPF study reproduced identical significance calls.
76+Modules across both technologies
7+Targets carried end to end
10Collaborating institutions in five countries
Ways to work with us
License, partner or train.
Run the technologies yourselves, have us run the science, or build the skills in your team.
License the technologies
- Hosted by us, in your cloud or on your own servers, or on our cloud
- Every module on both technologies, plus Boltz models
Partner on a programme
Our computational scientists run the work, from target assessment to a nominated candidate list.
- Target identification, virtual screening and de novo design
- A complete parameter trail behind every decision
Train your team
Hands-on training for company and institutional teams, led by the scientists who build the technologies.
- Workshops on campus, online or hybrid
- From RNA-seq reads to a docked complex in one sitting
We work for
Universities, institutes and companies.
Academic, our global presence
Bharati Vidyapeeth Deemed University
Nagpur College of Pharmacy
Nagpur University
GH Raisoni University, Saikheda
King Saud University
University of Warsaw
University College London
University of Calcutta
Corporate
Get started
Bring us a target.
Tell us where your programme stands. A computational scientist will show how the technologies apply to your problem and, just as importantly, where they do not.
+91 97405 00475
Bengaluru, India
What happens next
- 01Share your objectives
A short conversation about your target, data and timelines.
- 02See it applied to your problem
A tailored demonstration of the most relevant modules.
- 03Agree the right engagement
Licensing, a collaborative programme or training, scoped to your needs.