AlexGW (@ALGW71) header
AlexGW (@ALGW71) profile picture

Immunoinformatics Researcher.

Oxford, England
Joined January 2023
Napkin math on why your percentile rank thresholds for CD8+ T-cell epitope prediction are comically lax. 30mb exome -> let's say half is CDS -> 5M aa possible proteome. Cut those up into 8-11mers -> 20M peptides x 6 distinct MHC alleles -> 120M distinct possible pMHCs (cont)
6
7
45
15,071
Beautiful work by my colleague at OPIG immunoinformatics @ALGW71 -- on the ability of deep learning models starved of knowledge (e.g. an antibody canonical class) to extrapolate its existence from other training datapoints.
2
4
490
My colleage @AlissaHummer & I are organising a conference on Adaptive Immune Receptors: modelling and immunoinformatics, taking place this April 5th at Oxford University. Invited Talks - Posters (+selected talks) - Networking More info and registration: mgms.org/WordPress/conferenc…
1
14
31
3,060
Really happy to share our new study: Investigating the ability of deep learning-based structure prediction to extrapolate and/or enrich the set of antibody CDR canonical forms. 1/N
Investigating the ability of deep learning-based structure prediction to extrapolate and/or enrich the set of antibody CDR canonical ... biorxiv.org/cgi/content/shor… #biorxiv_bioinfo
1
4
21
1,846
The tools used to perform this study (ABodyBuilder2 and OAS) can be found on our website: opig.stats.ox.ac.uk/resource… 10/N
1
118
Massive thanks to @brennanaba and Prof Charlotte Deane and the whole OPIG team! Hope people find this interesting, and feel free to get in touch with any questions. Looking forward to presenting this work at #AET2023 on Wednesday this week! 11/11
2
1
398
Investigating the ability of deep learning-based structure prediction to extrapolate and/or enrich the set of antibody CDR canonical ... biorxiv.org/cgi/content/shor… #biorxiv_bioinfo
3
1
12
3,764
New small molecules preprint: Buttenschoen et al. present PoseBusters, a python package for measuring the physical plausibility of docked poses, and apply it as an additional quality control metric in evaluating a suite of recent docking methods. arxiv.org/abs/2308.05777
5
23
2,223