Zamin Iqbal
@ZaminIqbal
Followers
9K
Following
42K
Media
927
Statuses
48K
Professor of Algorithmic and Microbial Genomics at the University of Bath. Pathogens, genomics, genome graphs, antibiotic resistance, algorithms,data structures
Bath, UK
Joined November 2013
I'm going to be advertising a PhD studentship (open to international students) working on long read pangenome / assembly algorithms in bacteria. Would suit someone with maths/compsci background. Aptitude and interest more important than experience. Advert to follow. Pls RT!
1
64
68
Another thing that is maybe less emphasized in this paper is that CLINVAR is a great database of curated pathogenic/benign variants but it is extremely biased (in all sorts of ways) & should never be used as a representative benchmark dataset for most types of variants. 1/
Latest genomic AI models report near-perfect prediction of pathogenic variants (e.g. AUROC>0.97 for Evo2). We ran extensive independent evals and found these figures are true, but very misleading. A breakdown of our new preprint: š§µ
1
21
165
Very glad to see this preprint by @lh3lh3 and @meyersonlab confirming our finding of artifactual fold-back INV in long reads (Fig S1 in our @naturemethods paper presenting #SAVANA, which filters such artifacts to improve SV calling, hence their relevance https://t.co/AdZhqHUsDb
Detecting Foldback Artifacts in Long Reads https://t.co/YMYx1ilaOa
#biorxiv_bioinfo
3
10
57
Most people donāt get that āHarvardā isnāt one thing, itās three: a fictional college, a shorthand for elitism, and a real school with smart, dedicated people The administration thinks they are attacking the first, justifying it with the second, and itās the third who get hurt
13
22
158
Carl Sagan was really ahead of his time. Absolutely prescient quote from one of his final interviews.
0
3
54
Tomorrow at the Systems Virology Journal Club, @poly_microbial will present his beautiful work with @HarmitMalik on optimizing temperature-dependent trade-off between protein production and processing in alphaviruses through stop codon selectivity
pubmed.ncbi.nlm.nih.gov
Most mosquito-transmitted alphaviruses encode a premature opal termination codon upstream of their viral polymerase. We show that the Sindbis virus (SINV) opal codon outperforms other stop codons in...
Join us for the Systems Virology Journal Club! šStellar speakers will discuss fitness tradeoffs, superinfection exclusion, bacterial antiviral defenses, virome proteins, spatial transcriptomics, and adaptive and innate immunity. @wm_schneider Reg (free): https://t.co/16kOLKk2IP
0
4
8
šØ Just out in Science: 2/3 of bacterial gene families have a preferred position on the chromosomeāand natural selection put them there! https://t.co/svgKQZFiYy
@HHU_de @ceplas_1
5
68
251
The number of sequenced microbes is growing exponentially, but computational power grows at a slower rate. The result? BLAST ā the "Google of biology" ā can search only a fraction of sequenced microbes, and that fraction is shrinking exponentially over time. 2/
1
1
4
A decade ago, we had thousands of bacterial genomes. Now, we have millions. How to scale computational methods? Our paper in @naturemethods answers this: use evolutionary history to guide compression and search. ā¦From terabytes to tens of GBsā¦ w/@Baym @ZaminIqbal et al. š§µ1/
3
53
169
Thrilled that our work on this problem with @KarelBrinda, @ZaminIqbal, and others is out in @naturemethods today! We used phylogenetic compression (described in the thread) to compress every microbe ever sequenced onto a flash drive so that it can be searched with a laptop!
So we asked: what sets the fundamental limit on computation on large genomic databases? Evolution! The irreducible entropy in genome collections is bounded by the most parsimonious path to introduce that variability. In other words, optimal compression should echo phylogeny. 4/
3
29
134
BREAKING: Donald TrumpĀ has ordered a communications blackout at America's federal health agencies, per WaPo. The CDC, FDA, HHS and NIH have all been told to pause external communications, including publishing scientific reports, updating websites or issuing health advisories.
956
5K
28K
Update: He just got home and he doesnāt want to talk about it.
677
2K
194K
My kid came home from school yesterday and told me his class is supposed to dress up as Elvis for morning assembly. I told him I think he means elves. He disagreed. I canāt wait to hear how his day went when āThe Kingā returns home.
2K
18K
546K
csvtk users, if you use `filter2/mutate2/mutate3` a lot like me, please update it to v0.31.1. I've fixed the slow speed since v0.27.0 (Aug 2023)š
. Conda-forge and Bioconda are not synced now. please download the binaries from GitHub. https://t.co/sEeqcAwxgj
github.com
Changes csvtk v0.31.1 csvtk filter2/mutate2/mutate3: fix the slow speed, I was stupid before. #269 csvtk csv2json: further fix values with double quotes and new line symbols. #291 Notes...
1
5
17
Prof Kristie Dotson giving the @RIPhilo @cardiffuni annual lecture 2024 on epistemic frailty and the need for philosophical artists - hosted by our own @a_nonamename - exciting, confounding, inspiring š„
Sefydliad Brenhinol Athroniaeth Darlith Flynyddol Caerdydd 2024 Royal Institute of Philosophy Cardiff Annual Lecture 2024 āā :: Kristie Dotson :: āā Quiet As Its Kept: On Epistemic Frailty Tawel Fel Ei Gadw: Ar Eiddilwch Epistemig Live! Now! Byw! Nawr! https://t.co/U13jkFEy20
1
1
15
Many thanks to @ZaminIqbal @pmelsted and @Nicky_Mulder for a great meeting @eventsWCS. We met old friends, learned a lot, and made new connections for future research directions.
0
3
17
It is clear that @rfaure2 knows the cardinal rule of presenting on sequencing algorithms, you never talk about the reverse complement. #genomeinformatics24 Very cool work on reducing sequencing using sketches that are sequences and can be fed into other assembly algorithms
0
2
11
Next up, Caitlin Collins presenting "Molecular clocks in the accessory genome?: Investigating rates of gene gain, loss, and selection". #GenomeInformatics24
1
1
2
62.7% of total system energy is spent on data movement during processing. We want to lower energy costs by moving computation close to memory by moving compute to DRAM.
1
1
2
V interesting thread on using LoganSearch to find fusions
The output came as advertised in a few minutes. I got 2600 hits overall. That is more samples with TMPRSS2-ERG fusions than in the whole Cosmic database (2300). Wonderful! Although probably a good part are from the same few cancer cell lines.
0
0
10