Hugo Talibart

Computer scientist with strong interest in biology (proteins, homology search, (co-)evolution).

Currently postdoctoral researcher at ISYEB, MNHN Paris.

[Hugo Talibart @HTalibart]

Very excited to start my postdoc today at the Atelier de BioInformatique (ISYEB, Muséum National d'Histoire Naturelle). I will be working on improving our Potts model alignement method PPalign, hopefully making it possible to improve viral metagenome annotation.

[Hugo Talibart @HTalibart]

The paper describing our method for pairwise DCA Potts model alignment was published in BMC Bioinformatics!

"PPalign: optimal alignment of Potts models representing proteins with direct coupling information"
https://t.co/DoCLPaZO9s

[Hugo Talibart @HTalibart]

Only two days away from my PhD defense!
No buffet unfortunately but you're still invited to assist the live broadcast

[DyLiSS Team @DylissTeam]

Hugo Talibart (@HTalibart)'s PhD defense on "Comparison of homologous protein sequences using direct coupling information by pairwise Potts model alignments": 24th February 2021 14:30 (UTC+1). Follow at https://t.co/GCLRHv5dHk

[Hugo Talibart @HTalibart]

Many thanks to the #JOBIM2020 organization committee for their hard work on the organization of this virtual conference. I was happy to present our method ComPotts for optimal Potts model alignment. My slides for this morning's talk are available here: https://t.co/7B9PJ83SLI

[Hugo Talibart @HTalibart]

ComPotts, our method for optimal alignments of coevolutionary models for protein sequences, is now fully described on bioRxiv along with preliminary results: https://t.co/QTDBCUpx6r

See you at my talk @ #JOBIM2020 for more details !

[Nicolas Guillaudeux @nguillaudeux]

Hey ! Envie d'en savoir plus sur la recherche des doctorants ? Ça se passe le 17 octobre à partir de 20h au Diapason ! #Symbiose #Dyliss #Genscale #Rennes #SciencesEnCourts #Sciences #Recherche #VulgarisationScientifiques #TresCourtsMetrages #Doctorants

[Sarah Drasner @sarah_edo]

Please stop glorifying what's unnecessarily complicated. I've heard countless times people brag about how long it takes people to understand what they work on.

In my perfect world, it would be the opposite. You would brag about how quickly people can understand what you wrote.

[Grace @gracecondition]

everyone else using word2vec:
king – man + woman = queen

me using word2vec:
fish + music = bass
fish + friend = chum
fish + hair = mullet
fish + struggle = flounder
oink - pig + bro = wassup
yeti – snow + economics = homo economicus

https://t.co/UAiViuEnM2

[Viviane (Scilabus) 🇫🇷🇨🇦 @Scilabus]

Nouvelle vidéo ! Vous n’y avez probablement jamais prêté attention mais presque tous nos t-shirts ont un ourlet sur la manche. Il ne sert pas que à “finir la manche” ou à empêcher l’effilochement… non non, il sert à autre chose!
https://t.co/ohPo2Rc3lF
RT si vous ne saviez pas!

[Data Science Fact @DataSciFact]

If you anonymize your data before doing machine learning, your model might "learn" artifacts of the anonymization processes.

For example, if you make everyone's birthday January 1, the model might learn that people are only born on January 1.

[Mick W@tson ↙️ @BioMickWatson]

"Data were analysed with a custom Python script"

[Fermat's Library @fermatslibrary]

Every odd integer is the difference of 2 squares
Proof:
1) Pick an odd number
░░ ░░ ░░ ░░ ░░

2) Bend it in half
░░ ░░ ░░
░░
░░

3) Fill in the rest
░░ ░░ ░░
░░ ▓▓ ▓▓
░░ ▓▓ ▓▓

The odd nbr is the area difference of the big and small squares

[The Sound of Science @SoundofScFr]

Décision juridique de bloquer l’accès à Sci-hub et Libgen en France

https://t.co/GHJgME02nu

[Paul Graham @paulg]

If you care about your readers, do this for them: find the actual beginning of whatever you just wrote, and cut everything before it. It's ok to rub your hands before getting down to work, but don't do it by writing introductory text.

[Steven Pinker @sapinker]

1. Reverse-engineer what you read. If it feels like good writing, what makes it good? If it’s awful, why?

[Fermat's Library @fermatslibrary]

Kinesins are motor proteins that move molecules around the cell. They are powered by ATP and transport molecules along microtube filaments at a rate of 2000 nm/s, meaning they take 250 tiny steps a second. 🤓

[Michael Nielsen @michael_nielsen]

The determinant of a matrix is how much the matrix expands or contracts space. More precisely: it's the volume of the parallelepiped spanned by the columns of the matrix. (Ditto the rows.) This, btw, is why det 0 => non-invertible: one or more spatial dimensions has collapsed

[Sciences en Cour[t]s @scienceencourts]

[SaveTheDate&ThePlace] L'édition 2019 de Sciences en Cour[t]s va bientôt commencer !

Une réunion d'information et de rencontre avec les organisateurs aura lieu le mardi 18 décembre 2018 à 18h30 à... https://t.co/UJfe6G4eVw

[Unix tool tip @UnixToolTip]

One way to practice sed commands is to pipe echo out. Example: echo 'hello world' | sed s/world/orb/