Sign in

Aziz Zafar

@aziz-zafar.bsky.social
16 followers 36 following 4 posts

Ph.D. Student at Columbia University Biomedical Informatics | Machine Learning for Genomics. Aspiring scientific educator with a love for writing and parks.

PostsRepliesMedia
Aziz Zafar @aziz-zafar.bsky.social · 08/05/2026
Exciting news! Zack Pelland and I recently published our work on associations between non-concussive head impacts and gut microbiome changes in football players! You can find it on PLoS One: doi.org/10.1371/jour... Huge thanks to Dr. Ahmet Ay and Dr. Ken Belanger for their mentorship throughout!
doi.org
Non-concussive head impacts sustained during American football correlate with changes in gut microbiome diversity and composition
Non-concussive head impacts (NHIs) are a significant health concern among at-risk groups, including athletes and military personnel. NHIs are hits to the head or head acceleration events (HAEs) that d...
110
Aziz Zafar @aziz-zafar.bsky.social · 13/05/2025
How can we better understand pathogenic variants in intrinsically disordered regions (IDRs)? How do models such as AlphaMissense and ESM1b predict pathogenicity, when these regions typically exhibit lower genomic conservation than ordered regions? Read more: doi.org/10.1101/2025...
doi.org
Molecular dynamics simulations of intrinsically disordered protein regions enable biophysical interpretation of variant effect predictors
Predictive models for missense variant pathogenicity offer little functional interpretation for intrinsically disordered regions, since they rely on conservation and coevolution across homologous sequ...
122
Reposted by Aziz Zafar
Kevin K. Yang 楊凱筌 @kevinkaichuang.bsky.social · 30/04/2025
Protein language model likelihood are better zero shot mutation effect predictions when they have perplexity 3-6 on the wildtype sequence. www.biorxiv.org/content/10.1...
Relationship between perplexity and zero shot performance.
0113
Reposted by Aziz Zafar
Chao Hou @chaohou.bsky.social · 29/04/2025
Why do large protein language models like ESM2-15B underperform compared to medium-sized ones like ESM2-650M in predicting mutation effects? 🤔 We dive into this issue in our new preprint—bringing insights into model scaling on mutation effect prediction. 🧬📉
132