Parijat Biswas (@dewdroponsand) header
Parijat Biswas (@dewdroponsand) profile picture

Physicist seeking asylum in cell biology. Occasional artist, passionate gamer. Postdoc @LabLappalainen

Helsinki, Finland
Joined February 2022
Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)Photo by Parijat Biswas (@dewdroponsand)
Parijat Biswas retweeted
Many of my recent conversations with other neuroscientists revealed a common concern: the alarming proportion of essentially worthless systems neuroscience papers appearing in the top journals. Not exactly fraudulent (real fraud is actually not that common), but a familiar combination of weak data, arbitrary interpretations, and elaborate statistical or computational analyses that create an illusion of rigor. Basically highly polished garbage, with a fashionable topic and a narrative designed to appeal to editors. Of course, none of this is new, but it's getting worse and I don't think outsiders can even get an accurate picture of the latest systems neuroscience any more. As one colleague recently put it, if the public knew just how shaky much of the evidence really is, confidence in the entire field might collapse, and we might all go down if the public loses trust in the whole enterprise, especially now that AI seems to be improving so quickly. The deeper problem is systemic: our mechanisms for evaluating, rewarding, and propagating scientific results are broken. Scientists who produce reliable work are often marginalized. Their progress is considered too slow, and because of their perceived lack of productivity, often they can't even secure enough funding to keep their labs running. Some can't even get a job. I have seen many examples of this over the years. People with integrity are essentially punished for being meticulous. Meanwhile, the most relentless storytellers and opportunistic careerists accumulate funding, prestige, and influence, as they expand their academic families and place their people throughout the field. Criticizing a member in one of these "Corleone families" carries enormous professional costs. The result is not just a lot of bad papers. Entire research areas become saturated with unreliable findings and unsupported interpretations, reinforced by citation networks, press coverage, and awards. The longer this continues, the harder it will be to separate fact from fiction, and the greater the eventual damage to neuroscience itself.
60
160
40
849
61,517
Parijat Biswas retweeted
RT: Happy to share video & pdf of 1st course at College de France on Bioenergetics & Metabolism. This introduction course, from Vital heat To Free energy presents a history of thermodynamics in living systems. Video: urls.fr/VBYP7S PDF: urls.fr/H-gXeN Enjoy 🤩
3
12
65
2,509
Parijat Biswas retweeted
Phase separation driving chemical reactions!! This study shows how the environment inside liquid droplets makes polymerization easier and faster, allowing polymers to form spontaneously in water without a catalyst. pubs.acs.org/jacsat/article-…
18
127
6,310
Parijat Biswas retweeted
搞科研的都知道,画图比跑代码还折磨人。 数据跑完了,卡在配色、字体、排版和架构图上,能让人把头发全薅光。 今天挖到四个搞科研绘图的神器,建议直接放进收藏夹吃灰。 第一个,Scientific Visualization。 这玩意儿简直是期刊质检员附体。以前画个柱状图,总因为误差线没标 SEM 还是 SD 被审稿人怼。 它直接内置了顶刊规范,色盲友好和黑白打印模式全都兼容。 最关键的是,导出的 PDF、EPS 和 TIFF 直接满足分辨率要求,省去了反复返工的痛苦。 github.com/Panda-dda/Scienti… 第二个,Academic Research Skills 里的 Visualization Agent。 数据跑出来了不知道配什么图最合适,它能根据数据特点选图,并直接生成能运行的 Python 或 R 代码。 连 LaTeX 插图代码都一并给你写好,绘图和排版一步到位。 github.com/Imbad0202/academi… 第三个,Nature Figure。 专门治几张子图拼在一起像大杂烩的毛病。它会先问你核心结论是什么,再让每张子图各司其职。 生成后自动检查文字遮挡和边距对齐,对不齐直接不让你导出。 主打一个证据链清晰,让外行一眼看懂。 github.com/Yuan1z0825/nature… 第四个,深度之眼开源的 CCF-Figure。 AI 论文的架构图最让人头疼。它会先判断你的论文属于七类中的哪一种,再从十一种结构里匹配布局。 架构闭环和反馈分类都能安排,最后输出中英双语提示词,直接拿去给大模型出图,还附带了翻车自检清单。 github.com/Deepshare-Officia…
7
172
2
1,135
56,971
Parijat Biswas retweeted
pymol + blender had a baby and she said no to ai protein slop meet mol-blender from ur terminal / claude app / pymol baked w the legend @kinasekid !!! links in 🧵
9
79
4
588
24,639
Parijat Biswas retweeted
In their Review, Felix Campelo, Michael Krieg & colleagues discuss the imaging approaches advancing cellular mechanobiology from point measurements to spatially resolved maps of forces, material properties & viscoelastic fluid dynamics. #JCSImagingSI journals.biologists.com/jcs/…
4
14
900
Parijat Biswas retweeted
Please join us next Wed for Prof. Margaret Gardel's talk on "Cell size and mass density regulation in epithelia". For more info, visit cytophys.org
1
3
74
Parijat Biswas retweeted
Cellpose-SAM/Cellpose4: superhuman generalization for cellular segmentation github.com/mouseland/cellpos… vs Cellpose3 CellSAM microSAM SAMCell PathoSAM Omniposeac @marius10p @computingnature bioRxiv 2025 biorxiv.org/content/10.1101/…
2
34
192
11,596
Parijat Biswas retweeted
We’re making a personalized cancer vaccine for Yash, a 34 year-old pancreatic cancer patient, in the open with his consent. We want external experts to review our design and make their own contributions before we manufacture it at the RNA Foundry. RNAFoundry.com/yash
114
463
85
2,570
260,246
〜免疫染色Tips. 5〜 今回は膜透過処理で使う界面活性剤の種類についてです。 免染の際に細胞内部に抗体が届くようにするための膜透過処理。界面活性剤を使って細胞膜に穴を開けるのですが、ここで使う界面活性剤にも種類があることをご存知でしょうか? 最もメジャーなのはTritonX-100(0.1%程度で5~10分)かと思いますが、このTriton実は界面活性剤としてはかなり強いものに部類されます。細胞内への抗体の透過性は抜群に上がる一方で、ミトコンドリアなど膜構造をもつオルガネラを免染する場合には染まりが悪くなることも。 その場合は作用の穏やかなSaponin(サポニン)やDigitonin(デジトニン)、あるいは稀にTween-20も使ったりもします (いずれとも目安として0.1%前後で5~10分)。 以下にまとめましたのでご参考ください。 1. Triton X-100 (第一選択・強) 特徴: 細胞膜や核膜の脂質二重層を可溶化し、強力な抗体浸透性を与える。 注意点: ミトコンドリアなど膜構造をもつオルガネラの微細構造も破壊するため、綺麗に染まらない原因の1つに。 2. Saponin / Digitonin (弱・オルガネラ保存) 特徴: 膜中のコレステロールに結合して極小の穴を開けるため、ミトコンドリアなどの微細構造を保持したまま染めたい場合に有効。 注意点: Tritonに比べると抗体浸透性は落ち、コレステロールの少ない核膜には効きづらいため核内タンパクの染色には不向き。また、Saponinは作用が可逆的なため、膜透過処理以降の工程でも反応液に添加する必要あり。 3. Tween-20 (中・膜タンパク保存) 特徴: 主に非特異吸着の防止用に抗体反応液などに添加(0.1%程度)するのに使われる程良い界面活性剤ですが、細胞膜の膜タンパクを染める際にはTritonより良好な結果を出すことがあります。 【まとめ】 基本・大半のタンパク質 → Triton X-100 ミトコンドリア等のオルガネラ → Saponin/Digitonin 細胞膜の膜タンパク→ 低濃度TritonX-100もしくはTween-20 以上、TritonX-100を使っていてミトコンドリアや膜タンパクの染色がうまくいかない場合には、界面活性剤の種類を変えてみましょうというTipsでした。
1
27
1
293
19,563
Parijat Biswas retweeted
A thread of early experiments with Claude Fable 5.2: A 3D model of a cell, built entirely from code one organelle at a time. Built by @kevin_t_ngo.
484
1,333
295
17,646
1,442,815
Parijat Biswas retweeted
We've been calling it mitochondrial dysfunction for decades. Turns out that term is scientifically meaningless, and it's been holding back progress in understanding how mitochondria actually work. @MitoPsychoBio and colleagues published a perspective in Nature Metabolism arguing that the language mitochondrial scientists use fails to capture what mitochondria actually are: multifunctional organelles that dynamically recalibrate across dozens of independent activities depending on what cells need. The problem starts with oversimplification. Mitochondria get reduced to the powerhouse of the cell in introductory biology. That framing suggests a single primary job: make ATP. Reality is more complex. Mitochondria synthesize steroid hormones. They produce iron-sulfur clusters required for hundreds of proteins throughout the cell. They regulate innate immunity and antiviral responses. They control programmed cell death. They generate signaling molecules that reprogram nuclear gene expression. They can transfer between cells to rescue damaged tissue. Key insight: These aren't backup functions that kick in when energy production fails. They're simultaneous, independent, cell-type-specific roles that mitochondria perform constantly. Here's why the terminology matters: When researchers measure decreased ATP production and call it mitochondrial dysfunction, they're collapsing dozens of distinct mitochondrial activities into a binary working/broken framework. That creates problems: • A mitochondrion producing less ATP might be signaling rather than failing • Different mitochondrial activities can change independently • What looks dysfunctional in one context may be adaptive in another • Cell-type differences get lost when using generic terms The paper proposes a hierarchical terminology system with five levels: 1. Cell-dependent properties: Mitochondrial characteristics that vary by cell type. Neurons, muscle cells, and immune cells contain functionally distinct mitochondrial populations adapted to tissue-specific demands. 2. Molecular features: The structural and compositional elements. Membrane potential, protein composition, lipid content, mtDNA copy number, cristae architecture. 3. Activities: Specific biochemical processes. Oxygen consumption, ATP synthesis, reactive oxygen species production, calcium buffering, metabolite synthesis. 4. Functions: The biological purposes these activities serve. Energy provision, biosynthesis, signaling, quality control, cell death regulation. 5. Behaviors: Dynamic responses over time. Fusion, fission, movement, mitophagy, biogenesis, transfer between cells. The framework distinguishes what mitochondria are made of, what they do biochemically, why they do it biologically, and how they change over time. Example of how this improves precision: Old framing: This drug causes mitochondrial dysfunction. New framing: This drug decreases complex I activity, reducing ATP synthesis capacity while increasing superoxide production at complex III, without affecting calcium handling or steroid synthesis. The second version identifies which specific mitochondrial activities change and which remain intact. That matters for predicting side effects, understanding mechanisms, and designing interventions. The malleable nature of mitochondria creates additional complexity. Mitochondrial characteristics can shift within minutes to hours in response to nutrient availability, stress, or signaling molecules. These recalibrations are often reversible and physiologically appropriate. Calling a physiological recalibration dysfunction conflates adaptation with pathology. A muscle cell's mitochondria after endurance training look different from before training. Neither state is dysfunctional. They're optimized for different demands. The perspective argues this terminology shift will improve three outcomes: Teaching: Students learn mitochondria are multifaceted organelles rather than single-purpose powerhouses, building a more accurate mental model from the start. Research progress: Specific terminology forces specific measurements. Asking which mitochondrial activities change rather than whether mitochondria are dysfunctional leads to more mechanistic studies. Interdisciplinary collaboration: Clearer language helps researchers from different fields understand exactly what aspects of mitochondrial biology are relevant to their questions. The implications extend beyond semantics. How scientists conceptualize mitochondria shapes which questions get asked, which measurements get prioritized, and how findings get interpreted. Moving from function/dysfunction binary to multidimensional assessment requires more complex experimental designs. But that complexity better matches the biological reality. Mitochondria aren't failing when they recalibrate. They're responding. Understanding which specific activities change, in which cells, under which conditions, and with what biological purpose represents the path forward for mitochondrial science.
12
81
7
416
15,968
Parijat Biswas retweeted
Want to use AI for your next biological imaging project but don’t know where to start? This new free, online textbook from imaging experts at our Janelia Research Campus helps biologists & microscopists better understand & use AI for biological discovery: bit.ly/4sIPz5T
31
3
132
10,961
Parijat Biswas retweeted
My podcasts are way too long and way too technical. Few people have time to sit through 90 minutes of hardcore immunology! But if there's just one idea to take away from my interview with Jake Glanville at Centivax, a company developing universal flu vaccines and antivenoms, then I think it's this: First, influenza (and many other viruses or pathogens) mutate so rapidly that we keep getting infected by slightly different versions of them. And so the challenge of making a universal flu vaccine is to elicit antibodies that recognize many different strains, including ones that don't yet exist. The normal approach to this problem is just to mix together antigens from many different strains and inject them into humans all together. But your immune system then generates separate antibodies against each antigen, without necessarily producing antibodies that recognize ALL of them. (Imagine a vaccine with 20 different versions of influenza's hemagglutinin protein. Your immune system might make antibodies against version A, others against B, and others against C. But what happens when influenza mutates again, producing version 21? Your body has no antibodies.) What we really want are antibodies that recognize regions of the virus that rarely change! These so-called "conserved epitopes" are more likely to be shared across past, present, and future strains of ther pathogen. So Jake, noticing this problem, invented a thing called conservation–concentration coupling. The gist is that, instead of injecting ALL of the antigens into a human at high doses, you deliberately lower the concentration of each one. There's only a tiny bit of each antigen in the vaccine, in other words; too little to elicit an immune response on its own. (If a B cell comes along and recognizes only antigen A, that B-cell won't become strongly activated!) If, however, a B cell comes along that recognizes some converved region across MANY of the antigens, and it can bind to all of them, then it WILL become activated. Even though each individual antigen is present at a very low concentration, the B cell effectively encounters 20 times as much antigen, or 20x the normal dose. It is stimulated enough to proliferate, while the B cells recognizing only individual strains get "outcompeted." In other words, you're using antigen concentration to bias the immune system toward making broadly reactive antibodies. This is the key idea behind Centivax, and they've gotten really compelling data that this actually works in mice and humans (~180 patients in their phase I, which is now finished.)
13
27
7
158
11,101
Parijat Biswas retweeted
Thrilled to share our paper in @NatureComms RNA-binding proteins can enhance transcription largely independently of condensation Nascent RNA recruits FUS to transcription site, boosting RNAP loading and preventing R-loops to enhance elongation. nature.com/articles/s41467-0…
7
21
3
144
9,377