Our platform

Affinia Therapeutics is re-engineering the building blocks of gene therapy

ART: Affinia Rationally-designed Therapies

Our ART platform consists of three pillars with the aspiration to individually and synergistically improve the efficacy, safety, and manufacturability of AAV based gene therapies, including: (i) novel capsids with improved tissue targeting, immunologic profile, and manufacturability; (ii) novel promoters to precisely modulate transgene expression in specific cell types; and (iii) novel manufacturing approaches, for example, using new cell lines, plasmid designs, and transfection agents to improve quality, yields, and scalability. Collectively, we believe that our advances in these three areas will enable our gene therapies to achieve the desired benefit/risk profile and scalability to enable successful development of gene therapies in rare and prevalent diseases.

cART platform for novel capsids

We use complex computational algorithms, using structural modeling and mechanistic hypotheses, to rationally design libraries of novel AAV backbones and peptides.1 Our approach has enabled us to not only understand how changes in structure affect function, but also identify capsids with remarkably improved properties.

We design and manufacture these novel capsids and track them individually using unique barcodes. The capsids are then tested in experiments where we rank performance on various parameters that include tissue targeting and cell type specificity, magnitude of transduction and expression, packaging efficiency, immunological profile, and manufacturing yields. We generate terabytes of data on both the best and worst performing capsids using next-generation sequencing, which is the same technique used to map the human genome. Lastly, we apply complex analytics to these data, leveraging machine learning, to identify the top performing capsids and also to identify correlates between performance and underlying capsid structure. The more experiments we perform, the more robust the structure-function maps we develop.

pART platform for novel promoters

Promoters are an integral part of gene therapies. We believe that a limitation of current gene therapies is the use of non-specific promoters that are not optimized for expression in the desired cell types, leading to lower expression levels in the desired cells and the potential for adverse events due to expression in other cell types. We are developing the next generation of promoters. 

Our pART platform mirrors our cART platform. We computationally design libraries of novel promoters leveraging existing data on natural promoters and cell type expression profiles. We uniquely barcode each promoter so we can track it and make a gene construct with it. We then manufacture and test them in experiments, evaluating performance across various parameters and applying complex analytics on the datasets generated to identify the top performers. As we do with capsids, we develop structure-function maps, learning from both the strong performers and the poor performers, to hone our ability to rationally design future promoters. 

Image of two scientists conducting research in a lab

mART platform for novel manufacturing approaches

We believe that manufacturing approaches are as critical as efficacy and safety in gene therapy. We adapt the rational design approach to our manufacturing for each program, to match our clinical and commercial strategy, including such factors as dose, patient weight, route of administration, and number of patients. We aim to define and control an initial manufacturing process that we can sustain through the life cycle of each product candidate.

Our cART platform gives us early indicators of yield as one of the initial parameters we evaluate in capsid screening, and we have generated structure-function maps that may result in the systematic ability to design capsids with high yields. At every step of our research and development and in our selection of capsid, promoter, transgene, and product candidate, we focus on manufacturing quality, yield, and scalability.

We can’t wait to create what’s next.


  1. Zinn E, Pacouret S, Khaychuk V, et al. In silico reconstruction of the viral evolutionary lineage yields a potent gene therapy vector. Cell Rep. 2015 Aug 11;12(6):1056-68.